Geoscience Reference
In-Depth Information
be incorporated in an analysis cycle. Six-hourly and twelve-hourly analysis cycles
are combined with short-term forecasts, so that the cut-off time can be delayed and
operational products can be made available earlier, as well. A further characteristic
is the higher temporal resolution of 3h that makes use of short-term forecasts.
The use of these data provides some potential for improvements since the known
westward propagating waves can be well captured avoiding the need to propagate the
S
2
(
standing wave by interpolation. Instead we are able to consider both migrating
and non-migrating tidal components. Furthermore, the sampling data permit the
proper determination of the
S
1
(
p
)
p
)
,
S
2
(
p
)
, and
S
3
(
p
)
atmospheric tides.
pressure
tides using the annual mean model described in Ray and Ponte (
2003
). Then, sine
and cosine amplitudes of each model were convolved with the Green's functions
to determine sine and cosine amplitudes of the
S
1
(
We developed a global gridded model of the
S
1
(
p
)
,
S
2
(
p
)
, and
S
3
(
p
)
tidal loading
displacements in vertical and horizontal directions. In the convolution step, we did
not invoke the IB assumption but instead considered that the oceanic response at
subdaily timescales to the tidal variation in pressure is negligible (Tregoning and
Wa t son
2009
).
Since the amplitude of the vertical tidal loading displacement is about three times
larger than that in the horizontal displacement, we only show the displacements in the
vertical direction in Fig.
7
. The displacement magnitude for
S
1
(
l
)
,
S
2
(
l
)
, and
S
3
(
l
)
reaches
1-2mm in low latitude regions, but decreases to negligible displacements at the poles.
The
S
3
(
l
)
and
S
2
(
l
)
tidal displacement shows weak latitude dependency and its amplitude is
about ten times smaller than those of the
S
1
(
l
)
)
and
S
2
(
)
l
l
vertical tidal displacements.
It is well known that the
S
1
(
)
atmospheric tide is dominated by large non-
migrating components with complicated spatial distributions (Haurwitz and Cowley
1973
; Dai and Wang
1999
; Ray and Ponte
2003
). This signal is susceptible to signifi-
cant diurnal boundary-layer effects over land masses and land-ocean boundaries. Dai
and Wang (
1999
) mentioned that the upward sensible heat flux from the ground due
to solar heating contributes significantly to the non-migrating component of
S
1
(
p
.
The main migrating component is most apparent over the tropical oceans where the
progression of phases shows an approximately constant westward motion. These
S
1
(
p
)
tidal displacements,
where topographic and land-ocean boundary features are clearly seen.
The latent heating associated with convective precipitation, which has a strong
diurnal cycle and supplements the direct solar radiational heating, was found to
be important mostly for the
S
2
(
p
)
pressure tide characteristics are well captured in the
S
1
(
l
)
tide is
dominated by its migrating component, which is moving westward with the speed of
the mean Sun, and is regularly distributed over the globe (Dai andWang
1999
). These
S
2
(
p
)
tide. Therefore, oscillation of the
S
2
(
p
)
p
)
pressure tide characteristics can be seen clearly in the
S
2
(
l
)
tidal displacements.
tide has also been detected in the
temperature and wind fields in various radar and optical observations. The origin
of this tide is still uncertain. If it is a global and migrating tidal wave with zonal
wave number three, it is excited either by the third harmonic of heating due to solar
insolation absorption by water vapor and ozone or by non-linear interaction of the
migrating components of
S
1
(
According to Aso (
2003
), the ter-diurnal
S
3
(
p
)
)
and
S
2
(
)
. Interactions between the
S
2
(
)
p
p
p
tide and
