A regional tide model was preferred to correct both
datasets from the oceanic tide signal. It was developed by
NOVELTIS in the framework of the COMAPI project,
funded by the CNES . It is based on
a purely hydrodynamic solution of the TUGO-m barotropic
model (ex-MOG2D model) in which
altimeter data were assimilated. The resolution of the finite
element grid of the model reaches 500 m in the most costal
zones, to be compared with the 1=4 spatial resolution of the
global models. The comparison between the harmonic
analysis in Senetosa and the prediction of the COMAPI
model shows that they are very close and the model can
be trusted in the area (not shown here). Even if the model
is not perfect, the most important aspect is the tide differential
between the tide gauge and the altimeter points. In
this area, the spatial gradients of the tidal waves are very
weak and if the model contains a bias compared to the real
tide signal, it will appear similarly in the predictions for
both kinds of measurements and will consequently have
no impact in the altimeter bias estimates.