_____________________________________________________________________________________________________________________

- 1 - METOCEAN
AVAILABLE DATA






_____________________________________________________________________________________________________________________
- 2 - METOCEAN
Sea-states characterisation as well as assessment of climatologies are based on analyses of large databases. The three major
sources for these data are direct (in-situ) or satellite measurement and numerical modelling.

In-situ measurement :

In-situ measurement allows assessment of the main characteristics at the scale of the wave (Hmax, Cmax).
Main available information :

- time series
- directionality
- spectral content

Satellite sensing :

Instrumentation on-board satellite provides global information at the scale of the sea-state :

- significant wave Height
- directionality
- spectral content

Numerical wave models

Numerical wave models, whether they are used in forecast or hindcast mode provide the whole spectral content of the wave
systems.

_____________________________________________________________________________________________________________________
- 3 - METOCEAN
In-Situ measurement devices

Moored buoys

Method : wave characteristics identification, height, period and direction from the measurement of the dynamics of the float.
Dynamics can be evaluated either from { heave, pitch, roll} or { heave, surge, sway}. Dynamics is most of the time derived from
accelerometers records but other technologies such as GPS can also be implemented. Data : Height, period, direction

Limitation : data storage and transmission, sensitivity to the dynamics of the mooring, response to breaking, motion computation
(signal integrated twice).

Acceleration double summation induces filtering thus eliminates part of the information (second order components).

)] ( 2 cos[
8
2
) cos(
2
t
H
k t
H
e e q + =


Surface elevation measured by the buoy is obtained from measurement of the vertical displacement:

)] ( 2 cos[
8
2
) cos(
2
t
H
k t
H
z boue e e q + = =
After filtering it reduces to
) cos(
2
t
H
z boue e q = =


Crest elevation is obtained for
0 = t e


So that
8
2
2
H
k
H
+ = q
and
2
H
boue= q
Measured surface elevation is linearised.




_____________________________________________________________________________________________________________________
- 4 - METOCEAN
Spectrum estimate from buoys records

To facilitate the practical estimation of S(f,), it is common to decompose the spectrum into

( ) ( ) ( ) u u , , f D f S f S =

where S(f) represents the omnidirectional spectral density (in m
2
/Hz) obtained by integrating S(f,) over directions and D(f,)
denotes the directional distribution function (in rad
-1
), which is a function of frequency and direction and satisfies for any frequency
f the following condition of normalisation

( ) 1 ,
2
0
=
}
t
u u d f D

S(f) can be directly obtained from the heave motion using Fourier transform.
D(f,) will be obtained from measurement of the roll and pitch motions (or heave, surge and sway). This directional distribution
function, positive and 2-periodic, may be decomposed into the Fourier series

( ) ( ) ( ) ( ) ( ) | |

=
+ + =
1
0
sin cos
1
2
,
j
j j
j f b j f a
a
f D u u
t t
u

where the Fourier coefficients are computed as

( ) ( ) ( )
}
=
t
u u u
2
0
cos , d j f D f a
j

and
( ) ( ) ( )
}
=
t
u u u
2
0
sin , d j f D f b
j


_____________________________________________________________________________________________________________________
- 5 - METOCEAN
In order to obtain an accurate knowledge of this function, it is necessary to develop this series to a high order. In practice, only the
Fourier coefficients of 1
st
and 2
nd
order: a
1
(f), b
1
(f), a
2
(f) and b
2
(f) can be obtained by cross-spectral analysis of the heave, pitch and
roll time series which are assumed to stand for the elevation and slopes of the local free surface against the North (x>0) and West
(y>0) axes respectively. The buoy therefore records (over T seconds) the data {X
1
(t), X
3
(t), X
3
(t)} = {(t), d/dx(t), d/dy(t)} in the
(xOy) coordinate systems of the plane defined by the oceans free surface at rest, with x and y axes pointing towards East and
North direction respectively. As an alternative, the buoy may measure its own motions in that plane (surge and sway), so that the
cross-spectral analysis is applied to the displacement data {(t), x(t), y(t)}.
The most immediate recomposition technique consists in estimating D(f,) from series expansion up to the 1
st
or 2
nd
order,
according to the available data. Thus, the function D(f,) may roughly estimated as

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) | | u u u u
t t
u 2 sin 2 cos sin cos
1
2
1
,

2 2 1 1
f b f a f b f a f D
tfs
+ + + + =

This method is generally discarded because the resulting directional distribution often exhibits negative values as well as non-
physical secondary peaks.













Recomposition using the 4 first Fourier coefficients (after Tucker, 2001)


_____________________________________________________________________________________________________________________
- 6 - METOCEAN
Recomposition by parametric method

The precise knowledge of the directional spectral shape is tough to derive since a lot of information is missing when measuring the
directional properties of the wave field experimentally. Thus, the true shape has to be approached as best as possible from a
limited amount of field information. A simple way of estimating D(f,) using the only first order coefficients is by fitting a parametric
function for each frequency bin. Obviously, such a function imposes a particular shape to the distribution. In addition, sea states
made of two coexisting wave systems at close frequencies but distinct directions will not be properly reproduced after
recomposition for the spectrum will exhibit one peak only, to which the whole sea state energy will be attributed. However, at
particular locations where the sea states do not significantly vary in direction, such a method can be easily implemented to obtain a
relevant first guess of the spectral shape.
One usually uses the parametric cos
2s
function, defined as

( )
( )
( ) ( )
( ) ( )
( )
( )
|
.
|

\
|

+ I
+
I
=

2
cos
1 2
1
2
,

0
2
2
2 1 2
f
f s
f s
f D
f s
s c
f s
u u
t
u


where I denotes the Gamma function and s(f) the spreading coefficient, which can be estimated from the 1
st
-order Fourier
coefficients as

( )
( )
( ) f r
f r
f s
1
1
1

=


where
( ) ( ) f
c
f
r
1 1
=
with
( ) ( ) ( ) f
b
i f
a
f
r 1 1 1
+ =
, and
( ) f
u
0
is estimated by

( ) ( ) | | f c f
1 0
arg

= u

Similarly, expressions for both parameters can be derived from the 2
nd
order Fourier coefficients.
Other distributions, Gaussian, Poisson, von Mises and even bimodal distributions may be used instead of the cos
2s
formulation.


_____________________________________________________________________________________________________________________
- 7 - METOCEAN
Recomposition by Maximum Entropy Methods

The estimated Fourier coefficients a
1
(f), b
1
(f), a
2
(f) and b
2
(f) being given for each frequency bin, an infinity of directional shapes may
match this set of values. To find out a likely compromise, statistical methods were used, based on the concept of entropy, which
may be defined as a measure of uncertainty for a stochastic variable (H. E. Krogstad). Two integral formulations of the entropy
function were proposed, one by J . P. Burg and another one by C. E. Shannon. In both, the aim is to minimize or maximize the
functional, depending on the adopted definition.

( ) ( ) ( ) ( )
}
=
t
u u u
2
0
, ln , , d f D f D f D H
s
Shannon

( ) ( ) ( )
}
=
t
u u
2
0
, ln , d f D f D H
B
Burg

Burgs formulation leads to very narrow peaks and sometimes produces artificial double-peaks in unimodal sea states. However, it
provides a very good resolution for close directional peaks.
Shannons formulation is a bit more complex in terms of numerical resolution, and requires a robust iterative algorithm. In practice,
it avoids the creation of artificial double-peaked systems and therefore is more reliable for the automatic processing of large field
data. Let us stress that both Burgs and Shannons entropy formulations yield directional shapes that rigorously match the four first
Fourier coefficients, with distinct resulting shapes though. Accordingly, both are correct mathematically but not necessarily
physically.









_____________________________________________________________________________________________________________________
- 8 - METOCEAN

Methodology applies for various kinds of sensors measuring different
kind of parameter (slope, velocity, pressure.), providing a transfer
function between this parameter and surface elevation can be identified


























(After Benoit & al., 1997)
_____________________________________________________________________________________________________________________
- 9 - METOCEAN

Buoy surge, sway and heave time series Recomposed spectrum

_____________________________________________________________________________________________________________________
- 10 - METOCEAN


Waverider buoy Datawell






Wavescan buoy Ocanor


Buoys are generally used in networks, mostly in coastal areas for
continuous monitoring of sea-states. They can also be used for
shorter surveys for assessment of sites climatologies (coastal
management, resource, oil production).






_____________________________________________________________________________________________________________________
- 11 - METOCEAN

Need for quality check of wave data.





_____________________________________________________________________________________________________________________
- 12 - METOCEAN
French network CANDHIS : http://candhis.cetmef.developpement-durable.gouv.fr/



_____________________________________________________________________________________________________________________
- 13 - METOCEAN
USA : NDBC (Natiaonal Data Buoy Center) : http://www.ndbc.noaa.gov/rmd.shtml

worldwide network NDBC

_____________________________________________________________________________________________________________________
- 14 - METOCEAN
Rotor current meter

Rotor current meters allow for local measurement of current velocity.

Method : current velocity is derived from the angular velocity of a rotor spinning under the action of the flow. Direction is obtained
from compass measurement as the vane attached to the current meters aligns with the flow.
Current meters are often coupled with pressure, temperature and resistivity sensors.

Limitations : need for calibration, fouling of the mechanical parts, data storage, power.


_____________________________________________________________________________________________________________________
- 15 - METOCEAN
ADCP

Acoustic Doppler current Profilers (ADCP) initially developed for currents measurement, also allow monitoring of waves local
kinematics.

Method : Identification of the velocity field of particles in suspension based on Doppler effect. Coupled with pressure sensor and
distance meter.

ADCP uses Doppler shift for measuring the flow velocity. An acoustic pulse with known frequency f
e
is emitted. As it travels across
the water this signal can be backscattered by particles (sediment, plankton) in suspension in the water. If the particle has a
velocity v, the frequency of the backscattered wave is shifted according to the Doppler law :
c
v
f f
e Doppler
. 2 =
with c speed off the sound in the water.
Flow velocity can then be derived from the change in frequency of the measured backscattered signal, assuming that the particles
have the same velocity as the flow and that speed of sound is constant (measurement of temperature and salinity allow for
correction of this value). The elapsed time between emission and reception provides a measure for the distance of the scattering
body to the receiver so that averaging measurements over given ranges of particle locations along the transmission axis (distance
bins), it is possible to obtain profiles of flow velocity.
Resolution of high frequency ADCP allows for measurement of waves kinematics (horizontal and vertical velocity) from which a
surface elevation can be derived (linear theory).
Velocity fields measured at various depth allow estimation of surface elevation by applying linear transfer relations.
Information on flow direction requires multi-beam information.

) cos(
2
) , (
) sin(
] cosh[
)] ( sinh[
2
) cos(
] cosh[
)] ( cosh[
2
t kx
H
t x
t kx
kd
d z k Hgk
z
V
t kx
kd
d z k Hgk
x
V
z
x
e q
e
e c
c
e
e c
c
=

+
=
u
=

+
=
u
=

Data : current velocity, surface elevation, directional wave spectra
_____________________________________________________________________________________________________________________
- 16 - METOCEAN
Limitations : Calibration, data storage, power consumption, water quality

An interesting feature of the ADCPs is they can be mounted on the hull of a vessel for surveys.



Ifremer/dro/lpo




Current Meters mooring
_____________________________________________________________________________________________________________________
- 17 - METOCEAN
Pressure sensors

Method : measurement of static and dynamic pressures.
Data : Height, Period, direction if using network of sensors
Limitation : set-up, sensitive to fouling, use in shallow water
only (filtering of high frequency components)




t
P
c
c

u
=

( ) ) cos(
] cosh[
)] ( cosh[
2
, , t kx
kd
d z k Hg
t z x P e
+
=




High frequency components filtering










_____________________________________________________________________________________________________________________
- 18 - METOCEAN
HF Radars

HF radars are shore based remote sensing systems using the over the horizon radar technology to monitor ocean surface currents,
waves and wind direction. Technology based on Braggs wave backscattering.

Method : radars emit radio waves in the range 5 to 50 MHz.
The rough ocean surface interacts with these radio waves and
due to the Bragg effect backscattered signals can be detected
from ranges of more than 150 km. The Bragg effect describes
the coupling of the electromagnetic wave with the ocean wave
field. To fulfil the Bragg conditions the electromagnetic wave
length needs to have twice the wavelength as the ocean
wave,( e.g. for a 30 MHz radar signal with Lambda 10 m, the
corresponding ocean wave is 5 m.). Reflections from waves
that fulfil this condition will generate a dominant signature in
the received signal spectrum due to in-phase summation of
amplitudes. The expected signature is a Doppler shifted signal
with a specific Doppler Shift given by the velocity of the
gravity wave that fulfils the Bragg condition. These Doppler
shifted signals will be symmetrical around the normalised
centre frequency, as long as the ocean surface does not
move. An ocean current will shift these Bragg peaks up or
down in frequency. This additional frequency shift contains
the information used to calculate the velocity of the ocean
current.

Limitation :processing, limited to coastal areas
Accuracy for ocean current velocity depends on operating
frequency (fo) and averaging time. In order to obtain a
reasonable resolution it is necessary to average time series
over a few minutes.
Inversions algorithms for wave measurement require
improvement.
























Backscattered spectrum (WERA Radar)
Bragg Scattering
_____________________________________________________________________________________________________________________
- 19 - METOCEAN

HF Radar applications




























Map of surface currents Map of significant Wave Height
_____________________________________________________________________________________________________________________
- 20 - METOCEAN
Marine Operations Traffic control and support





A surface current map from the WERA HF-Radar during the EuroROSE
real time campaign in Feb 2000. The fairways to the terminals are shown in pale blue colour.
The current speed reaches almost 3 kn (1.5 m/s) in the most intense region of the Coastal Current.
Significant Wave Height (Fedje)
_____________________________________________________________________________________________________________________
- 21 - METOCEAN


Marine Operations Trajectory forecast for drifting objects
























Current Model Forecast HF Radar data


_____________________________________________________________________________________________________________________
- 22 - METOCEAN
Radars

Radars can be used to assess the directional spectrum of a sea-state.

Method : Bragg waves modulation by longer waves (sea waves) Estimation of the wind velocity (sea surface roughness)

Limitation : Processing is not straightforward, no direct information on waves, sensitivity to spray




_____________________________________________________________________________________________________________________
- 23 - METOCEAN
(Alcal U.)

_____________________________________________________________________________________________________________________
- 24 - METOCEAN

Visual observation

Visual observations of waves height and estimates of periods and direction by crews on-board vessels are logged and saved in a
database since 1949 (Global Wave Statistic - http://www.globalwavestatisticsonline.com/ ).
Additional scarce observations are available since 1854.

Limitation : variability due to observers criteria (night & day), limitation to the areas covered by the main navigation routes,
inaccurate estimate of the tail of the statistic distribution (sailors tend to avoid bad weather).





_____________________________________________________________________________________________________________________
- 25 - METOCEAN
Satellite Measurement

Various instruments embarked on-board satellites orbiting around the earth allow
measurement of a large number of parameters :

Color of the oceans
Surface temperature
Elevation of teh oceans surface (reference geoid, surface currents (slope), tides, local
elevations)
Surface roughness (wind, waves, oil spills, )



Avantages of the ocean space monitoring :

Whole earth coverage
Widespread measurement zones
Directional spectra


Limitations of the ocean space monitoring :

Time and space discretisation
Complex indirect processing methods
No information on individual waves
Quality of the measurements in extreme conditions (strong wind, harsh seas)





















J ASON-1
_____________________________________________________________________________________________________________________
- 26 - METOCEAN
Missions

Since the mid-70s (GEOS-3, 1975 SEASAT, 1978 GEOSAT, 1985) various programs (missions) for ocean space monitoring have
been developed.

TOPEX/POSEIDON (ESA, CNES, NASA 1992)
Altimeter radars

Programme ERS (European Remote sensing Satellite system)
2 satellites ERS-1 (1991-1996) et ERS-2 (1995)
Radar Altimeter
Scatterometer
SAR

GEOSAT Follow On (US Navy - 1998)
Altimeter Radar

QuickSCAT (NASA 1999)
Rotating scatterometer

ENVISAT (ESA - 2002)
Altimeter Radar
ASAR

J ASON-1 (ESA 2002)
Altimeter Radar
Radiometer




_____________________________________________________________________________________________________________________
- 27 - METOCEAN
Spatiotemporal coverage


Sampling step in time and in space (intertrace) of the data mostly depends on the orbital trajectory of the satellite, defined by its
altitude and inclination.



Mission GEOSAT TOPEX/ POSEIDON ERS-1 ERS-2 GFO J ASON-1 ENVI SAT
Altitude (km) 800 1336 800 800 880 1336 800
Inclinaison
(degrs)
107 66 98 98 107 66 98
Rptitivit
(jours)
17 9,916 35 (3 et 168) 35 17 9,916 35
Intertrace (km) 164 315 98 98 164 315 98



Choice for orbits results from a compromise between spatial coverage, sampling and sensitivity and power of the sensors. A large
number of potential perturbations is also to be taken into account (irregularities in the gravity field)

The coverage offered by a single altimeter, in both space and time, doesnt allow recording enough data for an analysis at the
synoptic scale. Typically, a satellite can totally miss one storm or only see a small part of it. Satellite data must then be combined
with other data sets or assimilated in numerical wave models.

_____________________________________________________________________________________________________________________
- 28 - METOCEAN

TOPEX/POSEIDON ground track (Repeatability 10 days Intertrace 135 km)
_____________________________________________________________________________________________________________________
- 29 - METOCEAN
Missions archived at CERSAT (http://www.ifremer.fr/cersat/en/welcome.htm )
ERS-1 & ERS-2

ERS-1 and ERS-2 are two twin earth observing satellites launched by ESA respectively in J uly 1991 and April
1995. As an element of the ERS ground segment, the CERSAT archives, processes and distributes the data of
the following sensors :
the RA radar altimeter (measurement of the sea surface topography, the wave height,...),
the AMI-Wind scatterometer (measurement of the surface wind speed and direction over the ocean),
the MW micro-wave sounder (brightness temperature, atmospheric vapour content,...).
The ERS-1 mission ended on 10th March 2000 (its sensors had been on stand-by since 1995). Nevertheless,
thanks to the length of the ERS mission (MORE than 10 years), an impressive continuous and homogeneous
series of sea surface parameters could be collected all along the mission.

ADEOS/NSCAT
NSCAT, onboard the J apanese satellite ADEOS, is a double-swath scatterometer (measurement of the surface
wind speed and direction over the ocean), built by NASA and launched on 17th August 1996. Compared to the
ERS scatterometer, this sensor provides MORE measurements, in another frequency band (Ku). Unfortunately,
due to a failure in its solar array, ADEOS was lost on J une 30th 1997.

QuikSCAT/SeaWinds
SeaWinds is a rotating scatterometer (measurement of the surface wind speed and direction over the ocean) onboard
QuikSCAT, launched by NASA in J une 1999. Thanks to its higher resolution and its wider sample pattern, value-added
products with higher time and space resolution, compared to ERS, can be generated. The launching of a new SeaWinds
scatterometer, onboard the J apanese satellite ADEOS II, is scheduled for November 2002.


_____________________________________________________________________________________________________________________
- 30 - METOCEAN
DMSP/SSMI
SSM/I is a microwave imager, onboard several American satellites of the DMSP series, measuring brightness
temperature over the sea surface. This is a passive sensor, which is used at the Dpartement d'Ocanographie
Spatiale together with other sensors to study polar sea-ice and ocean winds. Only radiometric data from F10,
F11, F13 and F14 satellites of the DMSP series are archived and distributed at CERSAT.

TOPEX/POSEIDON

The TOPEX altimeter, onboard the French-American satellite TOPEX/POSEIDON launched by NASA in
August 1992, provides information on the sea-surface height, the mean wave height and the wind speed
along the satellite track, covering the whole earth surface in 10 days. These data, received on CDROM from
the CLS/AVISO, are used at the Dpartement d'Ocanographie Spatiale together with other sensors for
calibration/validation, sensor drift monitoring, etc...
ENVISAT

ENVISAT is an earth observing satellite launched by ESA in march 2002. Numerous sensors are onboard
the satellite. CERSAT is especially involved in the calibration/validation of the Altimeter (RA2) and SAR in
wave mode (ASAR). The altimeter provides information on the sea-surface height, the mean wave height and
the wind speed along the satellite track. The SAR is intended to provide 2D Wave spectra (swell and wind
wave direction and density) thanks to an algorithm partly developped in the Laboratoire d'Ocanographie
Spatiale.

J ASON-1

(Agency : CNES & NASA)
J ason-1 is a joint french/american follow-on to the successful TOPEX/Poseidon mission. It provides precise
sea surface height measurements (within 2 cm accuracy) in order to support operational oceanography at
modelling and forecasting ocean circulation, currents, water temperature,... and to monitor sea level change
at climatological scale. Sea surface height, wave height or wind speed can be measured by the onboard radar
altimeter, complemented by additional instruments for improved atmospheric and orbit correction.
_____________________________________________________________________________________________________________________
- 31 - METOCEAN
Programme Sentinel

ESA is developing five new missions called Sentinels. These missions carry a range of technologies, such as radar and multi-spectral
imaging instruments for land, ocean and atmospheric monitoring.
- Sentinel-1 is a polar-orbiting, all-weather, day-and-night radar imaging mission for land and ocean services.
The first Sentinel-1 satellite will be ready for launch in spring 2014.
- Sentinel-2 is a polar-orbiting, multispectral high-resolution imaging mission for land monitoring providing, for
example, imagery of vegetation, soil and water cover, inland waterways and coastal areas. Sentinel-2 will also
deliver information for emergency services. The first Sentinel-2 satellite is planned for launch in 2014.
- Sentinel-3 is polar-orbiting, multi-instrument mission to measure variables such as sea-surface topography,
sea- and land-surface temperature, ocean colour and land colour with high-end accuracy and reliability. The first
Sentinel-3 satellite is planned for launch in 2014.
Sentinel 4 and 5 should be launched in 2019 as a continuation of the sentinel programme.

Sentinel missions will support the Copernicus Marine Service which aims to deliver information on the state and dynamics of the
ocean to help protect and manage the marine environment and resources more effectively.
_____________________________________________________________________________________________________________________
- 32 - METOCEAN
Altimeter

Method : radar emits pulses of electromagnetic waves towards the oceans surface. The slope of the power signal of the
backscattered wave, measured by the altimeter provides an estimate of the variance of the sea surface, hence of the significant
wave height

Incident wave front






Reflexions at the surface of the oceans

















backscattered power
Time (ns)
pulse
Satellite altitude
0 20
Hs1
Hs2
Hs2 > Hs1
_____________________________________________________________________________________________________________________
- 33 - METOCEAN
Measurement requires the position and altitude of the satellite to be known
accurately. This can be achieved with an accuracy of about 2 to 3 cm.

Various corrections can be made to take into account perturbations
occurring during electromagnetic waves propagation (humidity,
ionisation,)

Additionally, signals being often perturbed by various noises, significant
wave heights are evaluated by averaging measurements of large numbers
of pulses.
Pulses are emitted at a rate higher or equal to 1 kHz. Hs are then calculated
at a rate of 10 Hz to 20 Hz and finally averaged at 1 Hz.

Ground track of the wave depends on the opening of the beam and the
satellites altitude.

Because of the velocity of the satellite orbiting in space, distance on the
ground between two successive measurement at 1 Hz rate is about 6 to 7
km.

The order of accuracy on Hs measurement is:

Sup(0.5 m rms, 10% Hs)

It is also possible to assess a mean period defined by :

( )
4
0
4
1
m
m
Talt =














_____________________________________________________________________________________________________________________
- 34 - METOCEAN
Calibration of altimeters by inter-comparison with buoys data is also necessary.

Colocalisation criteria in time and space are requested for the inter-comparison process. For instance for comparison of altimeter
with buoy data, one should average measurements of 10 footprints (~60 km) centred on the buoys location and compare with
SWH obtained from a 30 minutes wave elevation time series provided by the buoy.



















Correction table: (after Queffeulou, Long term quality status of wave height and wind speed measurements from satellite altimeters , Proc, ISOPE 2003, Honolulu)

Hs cor=a*Hs sat +b
a b
ERS-2 1.074 -0.0079
GEOSAT FO 1.0802 0.0392
POSEIDON 1 0

J ul1998 Aug1998
1
1.5
2
2.5
3
3.5
4
4.5
Date
Hs (m)
Bouee
Topex
Ers2
Hs Satellites (m)
H
s

B
u
o
y

(
m
)

_____________________________________________________________________________________________________________________
- 35 - METOCEAN

Physical variability of the oceans
Evolution of significant Wave height along the track of Topex




























-40 -30 -20 -10 0 10
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Latitude ()
Hs (m)
Hs
interpolatio
_____________________________________________________________________________________________________________________
- 36 - METOCEAN
Synthetic Aperture Radar (SAR)

Method : Based on the Bragg waves backscattering principle. Brag waves backscattered energy is modulated by waves and
proportional to sea surface roughness.
Synthetic aperture : Correlation of the backscattered signals as the satellite travels allows improvement of the radars resolution.

SAR Applications are multiple :

- Land (Topography/Mapping,Vegetation,Soil),
- Water (Water Management),
- Ocean and Coast (Coastal Geomorphology,Ocean
Waves,Ship Traffic,Ocean Currents and
Topography),
- Snow and Ice,
- Atmosphere (Winds)
- Agriculture (Crop Yields,Forestry),
- Natural Disasters (Flood,Earthquake/Volcano,Oil
Slick,Landslide),
- Solid Earth (Tectonics/Seismic Activity)

SAR can operate in various modes according to the needs. In
wave mode , 5 km x 5 km imagettes are generated with a
step of 200 km. Complex inversion procedures allow estimate
of directional wave spectra from these data. Information on
wind and current surface can also been obtained from analysis
of sea surface roughness.

Limitations : High-frequency cut-off (For Single Look Complex
(SLC) SAR products, the limit is about 60m in wavelength and
for ASAR medium resolution products, the limit is around
175m ) no information on wind-sea, wrong estimate of
global Hs.

Still, SAR provides interesting information on swell
propagation direction compared to other spaceborne
instruments.



















5 km
5 km
Imagette SAR
Inversion process
_____________________________________________________________________________________________________________________
- 37 - METOCEAN

Swell retrieval near California such as using ENVISAT ASAR Single Look Complex product


Extracted SAR wave spectrum. Collocated Wave spectrum such as measured by direction buoy.
_____________________________________________________________________________________________________________________
- 38 - METOCEAN





Sea surface roughness Corresponding SAR wind field

_____________________________________________________________________________________________________________________
- 39 - METOCEAN

Swell wave inversion over Katrina hurricane Corresponding wind inversion
_____________________________________________________________________________________________________________________
- 40 - METOCEAN



Mean radial surface speed in the Agulhas Current
derived from 329 ascending Envisat ASAR WSM
acquisitions
Europe Mean Wind Speed
_____________________________________________________________________________________________________________________
- 41 - METOCEAN


_____________________________________________________________________________________________________________________
- 42 - METOCEAN

Oil slick detection Marine Traffic control


_____________________________________________________________________________________________________________________
- 43 - METOCEAN

Ice Detection and monitoring
_____________________________________________________________________________________________________________________
- 44 - METOCEAN
Hindcast wave models

Hindcast wave models are numerical models developed for the computation of wave generation and propagation across the oceans.
The main forcing input to these models is the wind, but additional forcing such as currents can be also used. Dissipation of wave
energy is also accounted for, whether it is induced by bottom friction or white capping.

Even though often called hindcast models, these waves models can be used in both hindcast or forecast mode. The only difference
comes from the input data used to force the model. In forecast mode, input will be forecasted (extrapolated) wind obtained from
atmospheric models while in hindcast mode, simulation can be replayed afterwards, using data sets corrected by assimilation of in-
situ or satellite measurement. Forecast simulations cover periods of a few days to a few weeks while hindcast simulation can be run
for longer periods, up to ~40 years in order to build databases for statistics and climatologies.


Avantages of wave models :

Global ocean coverage
Embedded refined meshes for accurate local simulations
Operational use
Numerical modelling
Comprehensive spectral and directional information

Limitations of wave models :

Part of the physics described in the models is based on
empirical laws
Sensitivity to the quality of input databases (values and
discretisation in time and space of the wind and current
fields).
_____________________________________________________________________________________________________________________
- 45 - METOCEAN
We consider here sea-states characterised by their energy (or variance ) spectral density F. It is assumed here that the scales of
variations of the currents and water depth are longer than those of the waves.

We associate to a wave propagating in a current having a velocity U the relative frequency o such that :
kd gktanh
2
=
o
(where g is the gravitational acceleration, k the wavenumber associated to the wave and d the water depth.)

and the absolute frequency
e
:
U k. + =o e


The energy spectral density F(k,o,e) characteristic of the sea-state evolves in time and space, so that : F(k,o,e, x,y,t)

Under the assumption that the current is non steady and non homogeneous,
e
varies and the energy of the spectral components
is not conserved because of energy transfers. It can be shown that the action density defined as
o
E
A , where E is the energy will
be conserved in such a non homogeneous flow. Hence, propagation of wave fields are modelled using the spectral action balance
equation ( )
( )
o
u
u
,
,
k F
k N = :

Which can be developed in the Cartesian form :


o
u
u
S
N N k
k
N y
y
N x
x t
N
=
c
c
+
c
c
+
c
c
+
c
c
+
c
c










Local rate of change of
action density in time
Propagation of action
in geographical space
Shifting of the
relative frequency
due to variations in
depths and currents
Depth-induced and
current-induced refraction
Source term
_____________________________________________________________________________________________________________________
- 46 - METOCEAN
with propagation velocities :

( ) u cos U
c
dt
dx
x
g
+ = =


( ) u sin U
c
dt
dy
y
g
+ = =


Source term S on the righthand side of the equation includes the various components of energy input and dissipation :

S S S S
S
bot ds nl in
+ + + =


S
in
: Wind-Waves interaction wind energy transfer to waves
S
nl
: wave-wave interaction, energy transfer within the spectrum (non-linear)
S
ds
: breaking and turbulence dissipation
S
bot
: bottom friction dissipation

Various formulations exist for these different source terms, mostly based on empirical or semi-empirical approaches and proposed
in the literature by various authors.

The usual formulation for the input Wind-Wave term is in the form :

( ) ( ) u o u o , , BE A
S
in
+ =


in which the linear A and exponential B growth parameters depend on wave frequency and direction and wind frequency and
direction.

During wave growth non-linear wave-wave interactions cause energy transfer between wave components, inducing changes in the
spectrum during its evolution in time and space. This non-linear wave-wave interaction S
nl
can be approximated by different
_____________________________________________________________________________________________________________________
- 47 - METOCEAN
formulations. The Discrete Interaction Approximation (DIA, Hasselman & al.,1985) is applied in most of the models in use
nowadays (3
rd
generation models, developed in the 90s).

Wave energy dissipation in deep water S
ds
is dominated by viscosity effects and mostly breaking (whitecapping). In intermediate
depth and shallow water, it is necessary to add another dissipation term, S
bot
due to bottom friction.

For instance
( )
( )
( ) u o
o
u o ,
sinh
,
2
2
2
E
kd g
C S
bot bot
+ =
where
Cbot
is a friction coefficients that depends on the orbital motion.

The three main wave models in use nowadays are :

WAM (Wave Modelling group 1988-1991)

- homogeneous and steady currents,
- large to medium scales

WaveWatch III (NOAA/NCEP)

- non-homogeneous and unsteady currents,
- grids from 120 km to few 100m
- data assimilation
- some limitations in shallow water

SWAN (Delft University)

- non-homogeneous and unsteady currents,
- refined grids from 50m to 1 km
- shallow water


_____________________________________________________________________________________________________________________
- 48 - METOCEAN
Wind data

Forecast and hindcast wind data for input into numerical wave models can be obtained from various providers running atmospheric
models. Main wind data providers for forecast are :

European Centre for Medium-Range Weather Forecasts (ECMWF)

ECMWF is an intergovernmental organisation supported by 32 States which provides operational medium- and extended-range
forecasts and a state-of-the-art super-computing facility for scientific research. (http://www.ecmwf.int/ )


10-day 'Deterministic' forecasts of Mean Sea Level Pressure, of Wind Speed and Temperature at low levels, and of the Height of the
500-hPa isobaric surface.
_____________________________________________________________________________________________________________________
- 49 - METOCEAN

NOAA/National Center for Environmental Prediction (NCEP)

GFS atmospheric model (http://www.emc.ncep.noaa.gov/gmb/moorthi/gam.html )


_____________________________________________________________________________________________________________________
- 50 - METOCEAN

_____________________________________________________________________________________________________________________
- 51 - METOCEAN
Meteo-France

Meteo-France provides data from its atmospheric models developed at different scales.

Arpge : Global model provides forecast up to three days
ALADIN France : regional model centered on western Europe. Provides two and a half days forecast on a 10 km grid
AROME : refined grid (2.5 km) regional model for France



_____________________________________________________________________________________________________________________
- 52 - METOCEAN
For long term statistics and re-analysis, some long duration databases were built through various research programs.

ERA 40

The ERA-40 project was started in 2000 and funded by the European Union. ERA-40 is a re-analysis of meteorological observations
from September 1957 to August 2002 produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) in
collaboration with many institutions.

In addition to the 10 metre u and v wind components, a large set of parameters are made available in the database: 2 metre
temperature, ice surface temperature, mean sea level pressure, sea-ice cover, sea surface temperature,
Near real time blended winds

Also available are spectra and parameters obtained from a wave model at 6 hours time step on a 1.5 X 1.5 resolution grid.
Available wave data include : 10 metre wind speed, 2-D wave spectra, significant wave height and periods for global sea-state,
swell and wind-sea.

_____________________________________________________________________________________________________________________
- 53 - METOCEAN
HIPOCAS

The HIPOCAS project (Hindcast of Dynamic Processes of the Ocean and Coastal Areas of Europe) is a project funded by the EU.
HIPOCAS aims at producing a 40 years hindcast of wind, wave, sea level and current climatology for European waters and coastal
seas for application in coastal and environmental decision processes.
To describe the small-scale atmospheric response three regional climate models were forced with the large-scale 40 years
reanalysis. The data are provided on a 50 x 50-km grid. Circulation models were used in the North Sea, the NorthEast Atlantic
south of UK, including Azores and Canary Islands, and the Mediterranean Sea. Wave models based on WAM were used in those
areas and also in the Irish Sea. An horizontal resolution of 10-km and temporal resolution of 3 hours were used.

Quickscat

NASA's Quick Scatterometer (QuikSCAT) mission was launched in 1999. Mission ended in 2009. QuikSCAT scaterrometer provided
ocean surface winds at 10m height, on 12,5 km and 25 km resolution grids.


_____________________________________________________________________________________________________________________
- 54 - METOCEAN
Blended wind

To enhance the spatial and temporal resolutions of surface wind, the remotely sensed retrievals are blended to the operational
ECMWF wind analyses over the global oceans. The blending method aims to provide 6-hourly gridded wind speed, zonal
component, meridional component, wind stress and the corresponding components at global scale. The spatial resolution of the
resulting wind fields is 0.25 in longitude and latitude. The remotely sensed wind observations are derived from near real time
measurements performed by Seawinds scatterometer onboard QuikSCAT satellite and by the three Special Sensor Microwave
Imager (SSM/I) onboard DMSP satellites F13, F14, and F15. ECMWF analyses are available at synoptic time (00h:00; 06h:00;
12h:00; 18h:00) on a regular latitude-longitude grid of size 0.56250.5625. ECMWF analyses are interpolated in space and time
over each satellite swath occurring within 3 hours from the synoptic time. The differences are evaluated at eachscatterometer and
radiometer wind cell of about 0.25 resolution.


_____________________________________________________________________________________________________________________
- 55 - METOCEAN

_____________________________________________________________________________________________________________________
- 56 - METOCEAN


SAR/Buoy Comparison
_____________________________________________________________________________________________________________________
- 57 - METOCEAN


SAR/Buoy comparison
_____________________________________________________________________________________________________________________
- 58 - METOCEAN


Sea-states parameters comparison - Buoy Hindcast - SAR
_____________________________________________________________________________________________________________________
- 59 - METOCEAN

Sea-states parameters comparison - Buoy Hindcast - SAR
_____________________________________________________________________________________________________________________
- 60 - METOCEAN





Comparison of Hs measurement from altimeter and buoy (co-location < 0.5)



_____________________________________________________________________________________________________________________
- 61 - METOCEAN
EDF Ocean Model Yeu Island























EDF Ocean Model Les Minquiers

_____________________________________________________________________________________________________________________
- 62 - METOCEAN