Berth 17: correction to the recorder series

The Berth 17 recorder produced a recurring pattern with an apparent period of twenty-nine hours. We attributed it to an environmental oscillation independent of the harbor plant. A review of the instrument log shows that sampling a sixty-minute plant cycle at fifty-eight-minute intervals reproduces the observed period. We withdraw the claim of a second thermal tide.

The neighboring plant ran a sixty-minute cleaning cycle. The recorder queried the sensor every fifty-eight minutes. The first submission mentioned the recorder's schedule without identifying the plant cycle. We also lacked an independent record sampled more frequently.

The difference between the intervals is two minutes per acquisition. Thirty acquisitions return the recorder to the original phase of the plant cycle. Thirty acquisitions at fifty-eight minutes each span 1,740 minutes (twenty-nine hours). What appears to be a slow oscillation in the archive can therefore arise from the sampling of a faster one.

Changing the recorder interval to fifty-five minutes would produce an eleven-hour alias under the same ideal conditions. An actual environmental tide would not shift its period in response to a change in interrogation schedule. We have found no independent support for the proposed second tide.