#16 added TICON-4 data set. It synthesized the datums by computing 19-year harmonic predictions using the harmonic constituents. In theory, this should get close to the actual datums, but it's not a repacement for deriving datums from actual water level measurements.
This should be replaced by writing scripts that download the original GESLA-4 data and emperically derive datums from water level measurements.