The sea slope problem revisited

Discrepancies in sea surface topography based on comparisons between the results of steric leveling and repeated geodetic levelings have identified what is known as the ‘sea slope problem.’ This problem is actually twofold: (1) the sea surface relief based on steric leveling differs significantly from that based on geodetic leveling along several generally north-south coasts, and (2) successively propagated levelings between several widely separated tide stations indicate that the stationary sea slope seemingly has been changing with time, whereas differenced sea level means between these stations indicate that the sea surface relief has remained virtually invariant during the same intervals. Reexamination of the three reported discrepancies between geodetically and sterically determined sea slopes indicates that the Australian example is based on leveling of a quality inappropriate to the comparison. The discrepancy developed along the Atlantic coast of the United States is limited to the reach between Portsmouth–Hampton Roads, Virginia, and Key West, Florida, where the accuracy of steric leveling may be especially vulnerable owing to the dynamic effects of the Gulf Stream. Reconsideration of the example along the Pacific coast of the United States indicates that the various discrepancies are due chiefly to intrasurvey movement and resultant distortion of geodetically defined height differences between tide stations. Agreement between the results of steric and geodetic leveling along tectonically inactive north-south coasts devoid of strong boundary currents is generally good. This observation supports the conclusions (1) that any directionally dependent systematic error in geodetic leveling is measurably insignificant and (2) that where allowance is made for the possible effects of major boundary currents or intrasurvey movement during levelings between tide stations, the sea slope problem tends to vanish.