150
L. R. M. Maas et al.
reaches a magnitude of about 1.2 m. This betrays a local resonant phenomenon. In
stark contrast, East of Madagascar, the M2 tide is close to the nodal point situated
on Madagascar. Its surface elevation amplitude is diminished to 10 cm and it shows
large phase differences along the coast.
Altimeter observations are assimilated into ocean tide models [37]. These models
benefit from the additional constraints provided by the presence of ‘sea truth’ in the
form of deep-ocean tidal observations of pressure. Such contact measurements are
useful because altimeter observations do not fully cover the ocean surface. Deformations in the tidal potential, especially near coasts, as well as the presence of coherent
yet weak internal tides cause inaccuracies in the determination of surface tidal amplitudes and phases.
For this reason, an effort has previously been put into comparing altimetryderived tidal elevations to directly-measured tides [34], for which Bottom-Pressure
Records (BPRs) have been used. The latter are insensitive to mooring line motions,
and may thus provide reliable estimates of the surface tides. However, it turns out
that long-term BPRs at depths in excess of 500 m are rare. There appear to be only
about 80 deep-ocean pressure records that are at least 100 km apart and that last
longer than one year [34]. In the South-West Indian Ocean there are none. Previous shelf and deep-sea bottom pressure measurements [24, 41] have also been used
to investigate high-frequency contributions by linear and nonlinear (nonhydrostatic)
internal waves, that fall outside the scope of the present investigation.
In the present chapter, we will determine tidal amplitudes and phases of BPRs
at moorings deployed at depths in excess of 500 m along two transects in the
South-West Indian Ocean, on either side of Madagascar. This region, described in
section “Measurement Sites and Instrumentation”, is not covered in the previous set
of deep-ocean pressure measurements. Altimetry-derived tidal constants for this area
are therefore possibly influenced by the presence of the Madagascar and Mozambique coasts and shelves.
The pressure recorders are situated at both a suite of ‘bottom locations’ (for the
purpose of this study defined to be within 75 m from the bottom) as well as at ‘middepth’ sensors within the water column, typically positioned at 500 m below the
surface or deeper. Details on these instruments are given in section “Measurement
Sites and Instrumentation”.
In section “Time Series of Pressure Measurements”, all long-term pressure observations are converted to equivalent sea surface displacements and are subjected
to a harmonic analysis, yielding tidal amplitudes and phases. Prior to performing
the Harmonic Analysis, events showing blow-down of instruments, especially of
those deployed far above the bottom, have been eliminated. In section “Coherent
Surface and Internal Tides”, the harmonic amplitudes and phases derived from BPRs
are compared to each other and to altimetry-derived tidal constants. Subsequently,
section “Deep Versus Mid-depth Pressure-Measurements of Tides” compares tidal
constants derived from BPRs to those obtained from pressure recorders within the
water column. In general, within the water column, surface tides explain less of the
observed variance in the pressure time series. Part of the tidal variability is due to
internal tides. Away from the bottom, amplitudes and phases of the coherent internal
L. R. M. Maas et al.
reaches a magnitude of about 1.2 m. This betrays a local resonant phenomenon. In
stark contrast, East of Madagascar, the M2 tide is close to the nodal point situated
on Madagascar. Its surface elevation amplitude is diminished to 10 cm and it shows
large phase differences along the coast.
Altimeter observations are assimilated into ocean tide models [37]. These models
benefit from the additional constraints provided by the presence of ‘sea truth’ in the
form of deep-ocean tidal observations of pressure. Such contact measurements are
useful because altimeter observations do not fully cover the ocean surface. Deformations in the tidal potential, especially near coasts, as well as the presence of coherent
yet weak internal tides cause inaccuracies in the determination of surface tidal amplitudes and phases.
For this reason, an effort has previously been put into comparing altimetryderived tidal elevations to directly-measured tides [34], for which Bottom-Pressure
Records (BPRs) have been used. The latter are insensitive to mooring line motions,
and may thus provide reliable estimates of the surface tides. However, it turns out
that long-term BPRs at depths in excess of 500 m are rare. There appear to be only
about 80 deep-ocean pressure records that are at least 100 km apart and that last
longer than one year [34]. In the South-West Indian Ocean there are none. Previous shelf and deep-sea bottom pressure measurements [24, 41] have also been used
to investigate high-frequency contributions by linear and nonlinear (nonhydrostatic)
internal waves, that fall outside the scope of the present investigation.
In the present chapter, we will determine tidal amplitudes and phases of BPRs
at moorings deployed at depths in excess of 500 m along two transects in the
South-West Indian Ocean, on either side of Madagascar. This region, described in
section “Measurement Sites and Instrumentation”, is not covered in the previous set
of deep-ocean pressure measurements. Altimetry-derived tidal constants for this area
are therefore possibly influenced by the presence of the Madagascar and Mozambique coasts and shelves.
The pressure recorders are situated at both a suite of ‘bottom locations’ (for the
purpose of this study defined to be within 75 m from the bottom) as well as at ‘middepth’ sensors within the water column, typically positioned at 500 m below the
surface or deeper. Details on these instruments are given in section “Measurement
Sites and Instrumentation”.
In section “Time Series of Pressure Measurements”, all long-term pressure observations are converted to equivalent sea surface displacements and are subjected
to a harmonic analysis, yielding tidal amplitudes and phases. Prior to performing
the Harmonic Analysis, events showing blow-down of instruments, especially of
those deployed far above the bottom, have been eliminated. In section “Coherent
Surface and Internal Tides”, the harmonic amplitudes and phases derived from BPRs
are compared to each other and to altimetry-derived tidal constants. Subsequently,
section “Deep Versus Mid-depth Pressure-Measurements of Tides” compares tidal
constants derived from BPRs to those obtained from pressure recorders within the
water column. In general, within the water column, surface tides explain less of the
observed variance in the pressure time series. Part of the tidal variability is due to
internal tides. Away from the bottom, amplitudes and phases of the coherent internal