On perturbations of harmonic constants in the Thames Estuary

Abstract

A long series of Southend sea-level observations, extending over a period of 51 years (1929–79), is analysed using the harmonic method of tidal analysis. The harmonic method of tidal analysis and prediction works on the assumption that the relationship of nodal terms with their respective principal terms in the physical tide is the same as in the equilibrium tide. It is concluded that although this assumption is very useful in accounting for those terms which are not separable from a given length of observations it is not always fully valid. As a result a systematic error is introduced in tidal analysis and prediction. It is observed that small secular trends are present in amplitudes and phase lags of some principal tidal constituents. In the diurnal band, amplitudes of O1 and K1 are almost constant, but the phase lag of K1 is increasing and the phase lag of O1 is decreasing. In the semi-diurnal band, amplitudes of M2, S2 and K2 are increasing, while the phase lag of M2 is decreasing and the phase lag of K2 is increasing. The influence of these secular trends in amplitudes is to increase the tidal range and extreme levels. This is confirmed by increases in the range of tide in the last one and a half centuries. Statistically, no significance can be attached to trends in phase lags because standard errors are large. A comparison of old records with tides synthesized by modern constituents for London Bridge suggests that the phase of the tide was almost stable there from 1683 to 1835; then a large change occurred after 1835. Similar changes are noted in the tidal range. The nature of these changes indicates that they are partly due to increased input at Southend resulting from secular trends in tidal harmonics and partly due to man-made alterations: embankment, bank raising and deepening of the estuary.