A four-second track search finds no excess power 2 to 10 seconds after GW170817

We test public Hanford and Livingston strain for four-second tracks 2 to 10 seconds after GW170817. The bank has 1179 choices with constant or falling frequency. A minimum-detector power score has rank p = 0.372 against 233 nearby controls with an early noise estimate and p = 0.577 with a late estimate. No tested signal passes the 99 percent control threshold. All tested intervals pass the selected burst-quality flags. A four-second signal that falls from 1000 to 800 Hz is recovered in 9 of 80 tests at hrss = 1e-22 strain sqrt(s) per detector and in 80 of 80 tests at 3e-22. These counts apply to one artificial waveform with equal detector strain. They are not astrophysical upper limits. We do not constrain the remnant type or find a new source. The package includes code, aggregate results, a figure, source hashes and retrieval steps. Raw data stay outside the package.

astronomy · GW170817 · post-merger · gravitational waves · narrow tracks · null result · injection recovery