swarp - FTP Directory Listing - Nansen Environmental and Remote ...

piegazeInternet and Web Development

Dec 7, 2013 (4 years and 24 days ago)

241 views

Proposal title:


Ships and
Waves Reaching Polar Regions



Proposal

acronym
:


S
WARP





Revised version
June

2013



Funding Scheme:


Collaborative Project


Small and medium
-
scale
focused research project


Call identifier: FP7
-
SPACE
-
2013
-
1




(SPA.2013.1.1
-
06)


Name of coordinating person: Laurent Bertino, NERSC



List of participants:


No.

Participant organization name

Abbreviation

Country

1

Nansen Environmental and Remote Sensing

Centre (coord.)

NERSC

Norway

2

Institut français de recherche pour l'exploitation de la mer

Ifremer

France

3

Nansen International Environmental and Remote Sensing Centre

NIERSC

Russia

4

National
Environmental Research
C
ouncil

N
ER
C

UK

5

NAVTOR AS

(
S
ME)

NAVTOR

Norway

6

Institut des sciences de la mer de Rimouski

ISMER

Canada

7

University of Otago

UO

New Zealand

8

OceanDataLab (SME)

ODL

France



Table of Contents


SUMMARY

................................
................................
................................
................................
.........

5

1. Scientific and technical quality, relevant to the Topics adressed by the call

................................
...

6

1.1 Concept and

objectives

................................
................................
................................
...............

6

The main concept

................................
................................
................................
..........................

6

Overall objective

................................
................................
................................
...........................

7

Specific Objectives

................................
................................
................................
........................

7

1.2 Progress beyond state
-
of
-
the
-
art

................................
................................
................................
.

7

Waves
-
in
-
ice modelling

................................
................................
................................
................

7

Sea ice modelling

................................
................................
................................
..........................

8

Wave Modelling

................................
................................
................................
............................

9

Model integration

................................
................................
................................
........................

10

Integration into onboard navigation services and shore based contingency services

.................

11

Satellite
remote sensing of waves in open ocean and marginal ice zone

................................
....

12

1.3 S&T methodology and associated work plan

................................
................................
...........

15

Overall strategy of the work plan

................................
................................
................................

15

PERT Diagram

................................
................................
................................
.........................

15

Table 1.3a Overview of the work packages

................................
................................
.............

15

Table 1.3b Timing of the different WPs and their components

................................
...............

16

Table 1.3c Deliverables

................................
................................
................................
............

17

Table 1.3d Milestones

................................
................................
................................
..............

17

Table 1.3e Workpackage description

................................
................................
.......................

18

2. Implementation

................................
................................
................................
..............................

29

2.1 Management structure

................................
................................
................................
..............

29

Table 2.1a Meeting plan
................................
................................
................................
...........

29

Steering Committee

................................
................................
................................
.....................

30

User Group

................................
................................
................................
................................
..

30

Coordination with the Marine Core Services

................................
................................
..............

30

Responsibility for project deliverables

................................
................................
........................

30

Communication Policy

................................
................................
................................
................

31

2.2 Individual participants

................................
................................
................................
..............

32

Participant 1. Nansen Environmental and Remote Sensing Centre (NERS
C)

............................

32

Participant 2. Institut français pour la recherche sur la mer (Ifremer)

................................
........

33

Participant 3. Nansen International Environmental and Remote Sensing Centre (NIERSC)

.....

34

Participant 5: NAVTOR AS (NAVTOR)

................................
................................
...................

36

Participant 8
. OceanDataLab SaS (ODL)

................................
................................
....................

39

Scientific and technical complementarity

................................
................................
...................

40

Table 2.3a Expertise complementarities matrix

................................
................................
.......

40

European dimension of the project

................................
................................
..............................

40

2.4 Resources to be committed

................................
................................
................................
.......

41

Table 2.1b Summary of staff effort

................................
................................
..........................

41

Narrative description of costs

................................
................................
................................
......

41

NERSC

................................
................................
................................
................................
.....

41

Ifremer

................................
................................
................................
................................
......

42

NIERSC

................................
................................
................................
................................
...

43

NAVTOR

................................
................................
................................
................................
.

44

NERC

................................
................................
................................
................................
.......

45

ODL

................................
................................
................................
................................
.........

46

3. Impact
................................
................................
................................
................................
.............

48

3.1 Expected impact listed in the w
ork programme

................................
................................
.......

48

3.2 Dissemination and/or exploitation of project results, and management of intellectual property)

................................
................................
................................
................................
........................

50

3.2.1 Target user groups for dissemination of SWARP results
................................
...................

50

Shipping and ship routing

................................
................................
................................
........

50

Interface to GMES Marine and Atmosphere services

................................
.............................

52

Table 3.2a Planned GMES/MyOcean service provision to SWARP

................................
.......

53

The oil and gas industr
y

................................
................................
................................
...........

53

Fisheries

................................
................................
................................
................................
...

54

3.2.2 Dissemination activities

................................
................................
................................
.....

54

3.2.3 How will knowledge and IPR be managed?

................................
................................
......

55

3.2.4 Further exploitation beyond the

project period

................................
................................
..

55

4. Ethical issues

................................
................................
................................
................................
..

56

5. Consideration of gender as
pects

................................
................................
................................
....

56

References

................................
................................
................................
................................
..........

57

glossary

................................
................................
................................
................................
..............

60

Letters of support

................................
................................
................................
...............................

62




SUMMA
RY


SWARP will develop downstream services for sea ice and waves forecast in the Marginal Ice Zone (MIZ) in
the Arctic. Waves in ice are one of the most hazardous phenomena for vessels and industrial activities in the
polar seas, but there are presently no ser
vices providing any information
about either the waves

themselves

or their effects on the ice state (in particular the distribution of ice floe sizes)
. The monitoring and
forecasting systems developed in SWARP aim to become operational by the
middle

of the

project, filling a
gap in the present marine services of GMES.
A waves
-
in
-
ice model will be first validated then included
in

the forecasting services provided downstream of MyOcean (Arctic sea ice forecast) and as part of Prévimer
(global wave forecasts).

In addition to wave and sea ice forecast models, the project will develop satellite
observation methods for waves in ice and other ice properties in the MIZ such as ice concentration, ice types,
ice thickness, ice drift and ice edge configuration. Existin
g and new satellite observing systems, especially
SAR, optical and altimeter data, will be utilized for retrieval of waves and ice properties in the MIZ. The
project will
be exploited

operationally
using

data from future GMES satellites such as Sentinel
-
1.
The
project will integrate the
improved

met
-
ocean services in
to

state
-
of
-
the
-
art
technology
for
onboard
navigation and shore
-
based contingency

planning
.

The
maritime transport
user group is directly inv
olved in
the project
through

the participation of an innovative SME developing
the latter navigation and planning
softwares
.

1. SCIENTIFIC AND TE
CHNICAL QUALITY, REL
EVANT TO THE TOPICS
ADRESSED BY THE
CALL

1.1 CONCEPT AND OBJE
CTIVES

Th
e main concept

The Arctic is experiencing rapid and drastic changes in sea ice conditions, with innumerable consequences
for the environment and human activities.
While

sea ice conditions

i
n the Southern Ocean have not changed
significantly in the last dec
ades
,

human activities are growing
,

increas
ing the

risk
o
f accidents.
Operations
around Antarctica are also inherently riskier than in the Arctic due to isolation of the southern continent, and
the wilder conditions, with larger waves and more broken ice.
That is not to say Arctic activity is without
risk


indeed
, oil and gas exploitation
, mining prospects in Greenland

and Northern Canada
and shipping
endeavours are pressured by insurance companies to increase the level of safety
there

(
Brigham and Sfraga
2010
). Icebergs, sea ice, waves and ocean currents are crucial factors to consider
in order to achieve this
.

Despite the risks, t
ourism, adventure, shipping and research have intensified marine traffic both in the
Southern and Arctic Oceans and will contin
ue to do so in the foreseeable future. In a globalized world,
accidents and bad planning based on poor information can have enormous (and unpredictable) repercussions
on the environment, human lives, the economy and the society in general.
The mission of t
he GMES Marine
Downstream
Services is to provide quality information and forecasts about the state of the sea in support of
targeted
groups of users.

Marine forecasts are provided by numerical models that are constrained to the real
world by assimilating d
ata obtained in real
-
time. In polar areas, ocean and sea ice forecasts include ice
concentration (probability of presence), thickness, and velocity, ocean currents, sea surface temperature
(SST) and waves. In addition, operational ice charting services and

remote sensing products are providing
complementary information about ice types, ice drift, currents, waves and SST. Waves are of particular
interest since they can make the sea impractical for delicate operations, disperse oil spills, complicate search
a
nd rescue, sink ships or damage infrastructure. When they reach the ice cover, they can transform a uniform
and slowly creeping ice sheet into a collection of small ice floes colliding with each other and oscillating to
and fro on the propagating swell.
(S
hort waves are rapidly damped.)
If not anticipated, such events can have
considerable repercussions on activities, especially because they

often

depend on remote oceanographic
conditions and
can thus be difficult

to predict
if

only

the local weather

condit
ions

are considered
.



Figure
1
: The SWARP project aims
at filling the gap existing in the
GMES service
s

with respect to waves
in sea ice by 1) extending the validity
of wave and sea ice forecasting
systems in the marginal ice zone
, 2)
developing remote sensing methods.
As of today, there is no forecasting
service in GMES that is valid for the
gray zone where waves
-
in
-
ice
processes dominate.






Having said this, large waves can also be generated within a dilute ice cover, and in

recent summers ice
-
free
regions in the Arctic Ocean have been large enough to allow this to happen

(See
Figure 3
below for
an
example of the conditions in

s
ummer 2012)
.

This has caused massive areas of the Arctic summer cover to be
broken and to behave li
ke a marginal ice zone (MIZ)


so called because this type of ice has traditionally
only occurred at the edges of the ice cover.
Unfortunately, today’s marine forecasting capabilities are not
able to make predictions about such environmental conditions whe
re most offshore operations will take place
in the Arctic, which occur mainly in

the
MIZ.
The goal of this project is to
provide operational
waves
-
in
-
ice
forecasting
and monitoring capabilities in order to

produce a
new
downstream navigation service
for

sa
fer
shipping
in Polar Regions.



Overall objective

The overall objective of this proposal is to
extend
operational

downstream service
s

supporting
maritime
transport
safety
in the Marginal Ice Z
one:

extend
forecasts of waves in
to

ice
-
covered seas,


forecasts of sea ice in the
presence of waves

and remote sensing

of waves and sea ice conditions
.
The
results will enhance the GMES
downstream
services for waves and sea ice in Polar Regions. These are
increasingly needed to support
maritime transport (th
e target application) but also
offshore
operations, civil security, and coastal and environmental management in

both

the Arctic and Southern
Oceans.

Specific Objectives

1)

Develop a wave model for the MIZ.

2)

Develop a sea ice model that includes the effects of
waves in the MIZ.

3)

Develop satellite remote
-
sensing methods for wave observations
for use

in model validation and
forecast interpretation.

4)

Validate the improved models using remote
-
sensing and in situ data.

5)

Integrate the i
m
proved models into
an
onboard

navigation software,
NavTracker
.

6)

Demonstrate

results

to selected users and
ensure
the
sustainability of the service beyond the project
duration
.

1.2 PROGRESS BEYOND
STATE
-
OF
-
THE
-
ART

Waves
-
ice interactions have been omitted in global and regional models of

the ocean and waves because
such models were initially run at a resolution too coarse to represent the MIZ, the area at the periphery of the
ice pack where wave
-
ice interactions are the strongest. With today’s high
-
resolution numerical models, the
MIZ is
now
sufficiently resolved that relevant processes can be effectively simulated. However, ocean waves
and their effects on sea ice are not represented. The proposal will develop models and parameterizations that
better describe the physics of the MIZ and implem
ent them in two types of operational systems, namely ice
-
ocean and wave monitoring and forecasting systems. The approach outlined in this proposal builds on
existing operational systems and aims to extend their capabilities to a level where they can provid
e forecasts
of wave penetration in ice covered seas and of the effects of waves on sea ice properties. This effort involves
recent advances in
waves
-
in
-
ice modelling
, in
wave modelling
, and in
sea ice modelling

in the MIZ, most
of which have been contribut
ed by the consortium partners. Model developments will be supported by
satellite remote
-
sensing

developed for the monitoring of waves
-
in
-
ice and their effects on sea ice. These
results from the project will contribute to improving the GMES Marine Core Serv
ices in the Polar Regions.

Waves
-
in
-
ice modelling

Modelling the propagation of waves in ice
-
infested seas is a complex problem with a long history (
Wadhams
1973, 1986, 2000,
reviewed by
Squire 2007)
. This field has made significant progress during the last

decade
with the result that these models can now be applied in very realistic settings. Wave scattering occurring due
to various ice inhomogeneities (leads, cracks, floes, polynyas, keels, ridges, roughness) can be calculated
very precisely and wave atten
uation can be obtained when the ice profile is known accurately.
Vaughan et al.
(2009)

have devised a method to compute wave attenuation for a unidirectional ice profile composed of
adjacent segments of ice of different thicknesses.
Kohout and Meylan (2008
)

considered the rectilinear
passage of waves through a distribution of floating elastic plates representing a dilute marginal ice zone and
computed the corresponding wave attenuation. Recognizing the fundamental three
-
dimensional character of
wave propaga
tion,
Bennetts et al. (2010)

applied the scattering model to periodic clusters of floes
representing an idealized three
-
dimensional marginal ice zone.

Loss of wave energy can occur by many mechanisms

other than scattering, such as under
-
ice turbulence,
mic
ro
-
cr
acking, ice breaking, roughness and

viscosity. Consequently the existing scattering models are only
accounting for a portion of the losses. In the early stages of formation, ice takes the form of ice crystals
dispersed in seawater (frazil ice) at the
top of the water column. Depending on weather conditions, frazil ice
may

form pancakes of varying sizes and thickness
es

in wavy conditions
, or

it can be piled up into a thick
slurry which then freezes quickly. It can also
evolve into a thin transparent lay
er of nilas in calm conditions
,
but this is very rare except in leads deep in the Arctic.

When trying to simulate the propagation and
attenuation of gravity waves in this type of sea ice, viscous models are more appropriate. They treat ice as a
viscous flu
id layer floating on seawater (
Jenkins and Dysthe 1997
,
Jenkins and Jacobs 1997
,
De Carolis and
Desiderio 2002
,
Wang and Shen 2010
) and represent conditions typically found in the Southern Ocean, the
Greenland Sea or in polynyas. In general, the marginal i
ce zone is composed of mixed ice types and all loss
mechanisms should be considered. Viscosity has been embedded in
many
scattering models (
Squire et al.
2009
) but this field of research needs more attention

as its inclusion is simplistic at this time
.

Unfortunately, even the highest resolution satellite images are unable to resolve the small
-
scale features that
affect the propagation of waves and, even if the sea ice characteristics could be sufficiently resolved, the
computational cost would be enormou
s. Instead, to match the averaged representation of sea ice given by
climate models or satellite images (10
2

10
3

m
2
), scattering models are used in conjunction with ensemble
averaging (e.g. Monte Carlo) methods to provide attenuation coefficients as a func
tion of averaged sea ice
properties (floe length, concentration, thickness, roughness, crack density, etc.).
Kohout and Meylan (2008)

and
Bennetts et al. (2010)

have performed this exercise in idealized representations of a marginal ice zone
with encouragi
ng results. During the course of the WIFAR project,
Dumont et al. (2011)

combined averaged
results of
Kohout and Meylan (2008)

with a floe breaking parameterization and applied it on the output of a
sea ice model of the Fram Strait (3.5

km resolution).
Wil
liams et al. (under revision
, a,b
)

produced a
two
-
part
paper following on from this work incorporating the more recent attenuation model of
Bennetts and Squire
(2012)
. The predicted extent of the MIZ, defined according to the maximum floe size permitted by

the
presence of waves, corresponded quite well with what is usually observed in that area (
Figure 2
). This work
provides a method to connect small
-
scale wave scattering models with large
-
scale sea ice models
and

adequately represent the marginal ice zone.


Figure 2. The marginal ice zone as seen in Synthetic Aperture Radar (SAR) from the ENVISAT satellite (left)
and from a ship (right, GFD License 1.2). Many types of sea ice coexist in the MIZ, from frazil crystals
agglomerated in small thin cakes to floe
s formed of consolidated

ice
.

Sea ice modelling

State
-
of
-
the
-
art sea ice models render the mean dynamic and thermodynamic properties of the snow and ice
cover at a grid
-
cell size of many kilometres or tens of kilometres. Depending on the model, varying le
vels of
sub
-
grid scale details (such as the ice and/or snow thickness distribution, ice types, ice age, etc.) can also be
included. Sea ice models used in the MyOcean Marine Core Services (NEMO/LIM

used in the Global
Marine Forecasting Center (MFC) and dev
eloped at NERC
;

HYCOM/EVP

used in the Arctic MFC and
developed at NERSC
;

HIROMB models) are all based on the viscous
-
plastic (VP,
Hibler 1979
) or the
elastic
-
viscous
-
plastic (EVP,
Hunke and Dukowicz 1997
) rheology originally designed to represent the
plastic flow of compact ice and the formation of leads and pressure ridges. They all neglect the effects of
waves, which are extremely important in the MIZ. In this portion of the ice cover sea ice dynamics a
nd
thermodynamics are ill
-
defined,
causing a significant
deteriorati
o
n
in the

quality of sea ice forecasts in the
MIZ. With the gradual retreat of the

summer

sea ice from the Arctic and the
steady

increase of model
resolution, the MIZ occupies a larger mod
el area and its forecasting becomes increasingly important.

Interactions of waves and sea ice in the MIZ have been well observed since the MIZEX and SIZEX
experiments (
Johannessen et al. 1983, 1987, 1992, 1994
,
Martin and Becker 1987
) and conceptual models

have been proposed (
Røed and O’Brien 1983
,
Smedstad and Røed 1985
). Ocean waves fragment large floes
into smaller ones

and

make them collide with each other (
Martin and Becker 1987
),
so that

the large
-
scale
dynamical response of sea ice to atmospheric and

oceanic forcings

is no longer that of a viscous
-
plastic
material
. Collisional rheologies have been proposed to better simulate the behaviour of sea ice in the MIZ
(
Shen et al. 1986
,
1987
,
Feltham 2005
), but none of these models have been tested in realist
ic situations nor
included in ice
-
ocean coupled systems. Recently, as a contribution to the MIZ (Total E&P) and WIFAR
(Research Council of Norway and Total E&P) projects,
Dumont and Lisæter (2010)

have included the
collisional rheology of
Shen et al. (1987
)

in a sea ice
-
ocean model (NERSC/HYCOM) and tested it in a
high
-
resolution nested configuration of the Fram Strait. In the absence of an explicit representation of the
effects of waves, results could not be adequately evaluated, but simulations showed tha
t the combination of a
collisional rheology with an EVP rheology is possible and would definitely improve the predictions for ice
behaviour in the MIZ. By including the waves
-
in
-
ice model (WIM) of
Dumont et al. (
2011
)

in

TOPAZ,
the
MyOcean Arctic
system

(
Sakov et al 2012
)
,

a major step
has been
taken in that direction

by initial
simulations of the record melt in summer 2012
,
using input from
Ifremer (See
Figure 3

below
)
.

Finally, even though models simulate the main processes affecting ice thickness, they
are poorly validated
due to the absence of adequate data sets. This also explains the slow development of operational sea ice
models that can’t fully benefit from refined parameterizations based on observations. For this reason,
SWARP will take advantage o
f the capabilities of existing and new satellites to monitor the ice thickness and
help validate and refine parameterizations (see below for more details about satellite data).

Wave
M
odelling

Ocean wave forecasting is based on phase
-
averaged spectral model
s. Their accuracy critically depends on
the quality of forcing fields, including winds and sea ice, and on the accuracy of the parameterizations for
wind
-
wave evolution (generation, nonlinear evolution and dissipation). Numerical schemes can also be
import
ant for long period waves. The understanding of wave evolution processes has made important
progress in the last decade (
WISE Group 2007
) and recent numerical model developments have shown
important error reduction
in the significant wave height when compa
red with
satellite altimeter
(
typically
less than 10%
;
Ardhuin et al. 2010
). This can be attributed to a better understanding of wave dissipation
processes, either for swells (
Ardhuin et al. 2009
) or the breaking of locally generated wind
-
waves (
Ardhuin
et

al. 2010
).

Today the model errors in absolute terms are highest around the marginal ice zones. After the wind, sea ice
and icebergs are respectively the two most important
factors

affecting waves. Using iceberg statistics derived
from altimeter waveforms
(
Tournadre et al. 2008
), a first parameterization of Southern Ocean icebergs in
WAVEWATCH III has considerably reduced model errors (
Ardhuin et al.
2011
) but further developments of
this parameterization are needed. In particular, large errors near the ice

edge remain due to the crude
treatment of sea ice as a time
-
dependent mask that suppresses all waves (infinite losses).

Building on our experience with Boltzmann
-
type equations for wave scattering by bottom topography
(
Ardhuin and Magne 2007
), our goal is

to work from the waves
-
in
-
ice models (see above) to produce
scattering cross sections and adapt the numerical techniques already implemented in WAVEWATCH III for
wave
-
bottom scattering. Friction under the ice will also be included in a form similar to the

effect of air
-
sea
friction already used for swell dissipation (
Ardhuin et al. 2009
). Remote sensing observations will be used to
validate the models in the near field of the ice and on a large scale (see below). In the current version of the
global WAVEWA
TCH forecast in Prévimer, the sea ice mask is taken from a weather forecasting model
(ECMWF), which is based on the coarse resolution (25 km) SSMI/I ice concentration and remains persistent
when running in operational forecast. This mask does not contain i
nformation about ice types. For numerical
reasons, the wave forecasts stop at a conservative distance from the ice edge. In the proposed work, we plan
to use ice types extracted from both satellite data and sea ice models to improve the

representation of
i
ce

in
wave models, increasing the synergy between the different components of the project.





Figure 3. Maps showing
the effects of the 2012 Arctic summer storms waves in TOPAZ
.

Wave predictions
from the Global WAVEWATCH III model, are read in and allowed to propagate out of their original domain
of validity. The left hand figure shows the estimated significant wave heights

(note the significant wave
height is as high as 4 m)
, an
d the solid lines indicate the boundaries of pack ice (unbroken ice). The right
hand figure shows the maximum floe sizes produced as the result of waves causing breakage as they travel
into the ice. The solid lines are the same as in the left
-
hand figure a
nd correspond to a maximum floe size of
230
m
.

Note the large amount of broken ice after the passage of storms through the Pacific side.


Model integration

The models described above can be separated into two categories. The first one refers to operational systems
providing forecasts at a large scale, while the second category refers to models and theories that simulate
small
-
scale processes. Waves
-
in
-
ice for
ecasting capabilities will be developed by two modelling systems of
the first category, namely the TOPAZ ice
-
ocean monitoring and forecasting system
(
t
he Arctic
MFC
in
MyOcean
, the only system presently able to locate the ice edge at the accuracy of 50 km
thanks to advanced
data assimilation
)
and the third generation wave model WAVEWATCH III as in Pr
é
vimer. These two
systems receive forcing from a weather forecasting model (
Figure 5
). The proposed work aims at
incorporating models of the second category int
o each of these two systems by means of coherently
embedded lookup tables, parameterizations, or sub
-
modules.
Figure 5

illustrates how these different models
connect with each other in order to enhance the existing operational systems. Although the output
of the
updated systems can be related, their development is essentially independent from each other, mainly
because they are built on different numerical frameworks and it is not the scope of this proposal to fully
couple the wave and sea ice models togeth
er, but rather to extend their validity so that they can provide
complementary and improved forecasts in the MIZ. Both the WAVEWATCH
I
II and NERSC
-
HYCOM
models include routines for downscaling and
will

be used in nested configurations, using the same model

equations at different resolution. This simplifies the porting of the developments done from a local model of
the Fram Strait to another model of the Barents Sea or to a large scale model (pan
-
Arctic or global).


Before they become part of GMES Services,

model updates will be validated against high quality data, some
of which will be acquired and produced within the course of this project. Monitoring and measuring waves
and their effects on sea ice in polar environment represent a paramount challenge. For
this reason, this
proposal will build on existing initiatives and centres of expertise
(see institutions CVs in
Section 2.4
)
to
increase both the quality and the quantity of relevant data. Satellite remote sensing,
and
in situ data will thus
contribute to
the project through model validation,
and

also to the expansion of the dataset available to users
of GMES.



Figure 4
. Schematic of the proposed
model

developments.

Two operational forecasting systems (the ice
-
ocean model, and wave models) will be
extended by incorporating waves
-
in
-
ice models and
parameterizations. The satellite data will serve for parameterization, validation and also as part of the
future forecast services in the MIZ
.



Integration into onboard navigation services and shore based
contingency services

In the last 2 years, Navtor has changed the way “Paperless navigation” is working, and are now pushing
the
International Maritime Organization
IMO’s mandatory
electronic chart display and information system

(
ECDIS
)

requirement

(see
http://www.navtor.com
)
. By taking advantage of the “Pay As You Sail” principle,
Navtor has made use and updating of ENC (Official Electronic Charts) much more easy and less costly. It
has also removed obstacles and admin
istration related to weekly updating of ENCs. In principle the Navtor
ENC service has the following components: a Navtor data server, communication between ship and shore, an
onboard planning and updating tool (NavTracker
, see
Figure 5
), an optional
route
-
planning tool (NavPlanner


to come) and ECDIS functionality.

In SWARP Navtor plans to extend Navtor navigational services with a MIZ module, demonstrating data
harvesting from SWARP partners, extension of Navtor Data Servers and the onboard transmis
sion service,
and finally integration with the on board decision
-
and
-
contingency systems, NavTracker and NavPlanner.
NavTracker is today widely used by offshore operators, and also by short sea ship owners in Europe.


NavTracker today has two main function
s: it is an administrative tool for the ENC service “Pay As You
Sail”, and it is a contingency tool. All ship owners need to have a contingency tool dedicated to support a
vessel in an emergency all around the world, on a 24/7 basis. Basic needs are update
d with electronic charts
(ENCs), detailed weather and sea state information worldwide, tools for range and bearing, access t
o AIS
information, etc. However


m
ost importantly



t
hey are updated with the ability to poll positions on request.
Nav
T
racker’s co
ntingency module aims to fulfil all these requirements, and is currently used in ship owners
´

office and on board. Unfortunately, specific forecasts or observations in the MIZ are not available. For
offshore and shipping companies planning to operate in M
IZ regions, such a service is of crucial importance.
As commercial shipping groups plan for an Ar
c
tic shipping route, the request for more detailed ice
information will also increase. The outcomes of SWARP will bridge this gap between available and needed
information on the MIZ.



Figure 5: Screendump from NavTracker showing vessels’ positions and MetOc forecasts

in the Barents Sea
.


Satellite remote sensing of waves
in open ocean and marginal ice zone

Ocean waves have a direct effect on the scattering pro
perties of electromagnetic waves emitted by air
-

or
spaceborne radars. Short waves in equilibrium with local wind have the principal effect of roughening the
surface while longer waves modulate the surface velocity and local slope. Inversion techniques for

the
retrieval of long wave spectra from the observed modulation of velocity and slope on
Synthetic Aperture
Radar (SAR)

images were developed 20 years ago and are used for routine monitoring of swell in open
ocean using a dedicated SAR wave mode onboard ERS1
-
2 and ENVISAT SAR (see
http://soprano.cls.fr/L3/fireworks.ht
ml
) or in coastal zone using ERS2 and ENVISAT ASAR image mode
(
http://soprano.cls.fr/L2/waveProducts
).

Whereas the short waves are damped in the presence of ice on the sea surface, longer waves can prop
agate in
ice
-
covered ocean over distances of hundreds of
kilometre
s and therefore continue to modulate the local
surface velocities and slope. First observations of SAR signature of waves in the MIZ have shown the
potential of such measurements to study th
e evolution of wave parameters such as attenuation, dispersion
and refraction (
Wadhams et al. 1994, 2004
,
Vachon et al. 1993
,
Liu et al. 1991a
,
1991b
,
1992
,
Vachon and
Bhogal 1990
) but were limited to a very few SAR datasets. However, ERS1
-
2 and ENVISAT pr
ovide us
now with 20 years of SAR data over ice that could be used to revisit the initial analysis and possibly go
towards the systematic retrieval of waves
-
in
-
ice parameters from routine SAR observations over the MIZ for
NRT monitoring and model validatio
n purposes.

Satellite SAR images provide a unique opportunity to map wave propagation in the MIZ. SAR data from
airborne campaigns in the 1980s and ERS SAR data in the 1990s have been used for observations of ocean
waves in sea ice in several studies (
Lyze
nga et al. 1985, Wadhams and Holt 1991, Liu et al. 1991a, 1991b,
Vachon et al. 1993
). Different models were proposed to explain phenomena like wave refraction at the ice
edge. Whereas some studies consider hydrodynamic ocean
-
wave

sea
-
ice interaction alone
(
Shuchman and
Rufenach 1994
) others also discuss SAR imaging artifacts (
Vachon et al. 1993
). An ocean wave inversion
scheme for SAR images in the MIZ has been studied by
Schulz
-
Stellenfleth and Lehner (2002)

that
allows
the retrieval of wave spectra from S
AR images (
Figure
6
).



a


b

Figure
6
. (a) Example of wave propagation in the MIZ observed in an ERS
-
1 SAR image from
1 February

1992 in the Greenland Sea. The image covers 50 by 50 km and shows open water as
a
bright signature and
thin ice as

a

darker signature. (b) Wave spectrum obtained form the 5 by 5 km image taken in the same area
(
Schulz
-
S
tellenfleth and Lehner

2002
).


The SAR imaging mechanism is sensitive to both the radar cross section and the motion of the sea surface.
Models describin
g this complex mechanism and inversion schemes for the SAR retrieval of two
-
dimensional
ocean wave spectra have been developed (
Alpers et al. 1981, Lyzenga 1988, Hasselmann and Hasselmann
1991, Mastenbroek and de Valk 2000, Hasselmann et al. 1996
). In addition to wave observations, SAR
images can also provide information about the effect of waves on break
-
up of ice in the MIZ.
Sandven et al.
(1999)

identified a zone of broken
-
up ice floes in SAR images from the Barents Sea MIZ. This zone is
usuall
y identified
by a

brighter signature than the surrounding ice in the SAR images, and it is observed
repeatedly in areas where the MIZ is exposed to waves coming from large ocean areas.

However,

s
ince SAR
images have inherent speckle noise, it can often be
difficult to identify sea ice features in the MIZ using SAR
data alone. By using high
-
resolution optical images from aircraft photograph or satellite data, it is possible to
map the MIZ in more detailed as shown in
Figure
7
. For studies of sea ice conditio
ns in the MIZ it is useful
to include optical images as a supplement to the SAR images.


Figure 7
. (a) Aerial photograph of a 8 by 8 km area of the MIZ in the Barents Sea taken on 8 March 1992, a
few hours after the overpass of ERS
-
1 collecting SAR images

with resolution of about 30 by 30 m. (b) Subset
of the SAR image covering the same areas as the photograph in (a). The SAR image cannot discriminate the
ice bands from open water very well, but the wave propagation is clearly identified in the SAR image
(
Sandven et al. 1992)


In SWARP, SAR data will be collected in the field experiment areas in Arctic and Antarctic for analysis of
wave damping, refraction and penetration into the MIZ, as well as for mapping of the zone consisting of
small floes broken up
by waves. SAR data will be obtained from RADARSAT
-
2 and TerraSAR
-
X. Sentinel
-
1 images will provide extensive coverage of the area under study, but with improved spatial resolution
compared to ENVISAT data. Another important feature is availability of dual
-
polarization data (HH+HV or
VV+VH). It was shown that dual
-
polarization data provide better detectability of different ice types and
features (
Scheuchl et al., 2004; Yu et al., 2012; Sandven et al., 2012
). Particularly important is its possibility
to reliably identify sea ice and open water in the marginal ice zone. The resolution of the SAR images needs
to be of order 10 m to resolve the wavelengths of waves penetrating into the MIZ. In addition,
where
possible (in cloud
-
free daytime conditions),
high
-
resolution optical images will be obtained near
simultaneously with the SAR images in order to analyze the ice conditions in more detail. The images will
be analyzed for estimation of wave spectra and penet
ration into the MIZ. The wave spectra will be used in
combination with in situ measurements of wave periods and amplitudes to obtain the key wave parameters
for the wave model development.

To compute a transfer function from the SAR
-
derived wave spectrum t
o attenuation coefficients across the
MIZ, a detailed sea ice classification is needed. A sea ice charting service is offered in the Ocean and Sea Ice
SAF, in ESA’s PolarView project (http://www.polarview.org) and within the corresponding OSI TAC of
MyOcea
n. All are based on satellite data.
New downstream GMES services have started before SWARP,
like the SIDARUS project led by NERSC.
However, none of these services will offer a sea ice classification
product detailed enough to extract quantitative informatio
n on wave attenuation from the SAR data.
Different ice classification algorithms exist in the central ice pack (
Alexandrov and Piotrovskaya, 2008
) but
none in the MIZ.

As a perspective,
the
Sentinel
-
3 altimeter will
carry out
SAR
measure
ments of

freeboard

profiles in the same
way as CryoSAT and will therefore provide information on the propagation properties of waves
-
in
-
ice. In the
same time frame, a complementary CNES instrument SWIM will fly on a French
-
Chinese satellite CFOSAT
to estimate wave spectra a
t a global scale in
cluding possibly in the MIZ. Thi
s remote sensing instrument that
has already been planned will enable the operational implementation of the remote sensing component of
SWARP in the 10 years to come.

1.3 S&T
METHODOLOGY

AND ASSOCIATED WOR
K PLAN

Overall strategy

of the work plan

SWARP adopts a splitting of Work Packages (WP) by types of activities. Each WP is led by the expert in the
consortium that will continue exploiting the results after the comple
tion of the project. There are 4

resear
ch
and technical development (RTD) WPs in th
e project, one demonstration WP
6
, one dissemination WP
7

and
the
overall management WP
1

(
Figure 8
).
WP3

delivers ice parameterizations to modelling tasks both in WP
2

and WP
3
. The modelling

WPs

WP
2

and WP
3

communicate by an offline coupling, providing boundary
conditions to each other. Th
e observation

tasks in WP
4
, in addition to previous in situ and user data,

provide
parameters for the two modelling

WPs and feed the integration and validation in WP
5
. Once

the models and
sat
ellite data are validated in WP
5
, a user demonstration can take place
in WP
6
. The dissemination in WP
7


is targeted at users and aims at ensuring a sustainable service
.


PERT Diagram


Figure 8
.
PERT diagram showing the links between the

different work packages
.



Table 1.3a Overview of the work packages


No.

Work package title

Type of
activity*

Lead
part. No.

Lead part.
Short name

Person
months

Start
month

End

month

WP
1

Management

MGMT

1

NERSC

4

1

36

WP
2

Wave modelling

RTD

2

IFREMER

32

1

3
2

WP
3

Sea ice modelling

RTD

4

NERC

37.3

1

3
2

WP
4

Satellite remote sensing of waves in ice

RTD

8

ODL

46

1

36

WP
5

Integration and validation

RTD

1

NERSC

6
5

1
2

24

WP
6

Demonstration

DEMO

5

NAVTOR

22

3
0

36

WP
7

Dissemination and exploitation

OTHER

1

NERSC

19

1

36





Total

2
25
.3



*

MGMT: Management, RTD: Research and Technological Development, DEMO: Demonstration
, OTHER:
Other activity
. In accordance
with

the CP funding scheme.


Table 1.3b Timing of the different WPs and their components

Task

WP
title
/ Task
short
title

Month 1
-
12

Month 13
-
24

Month 25
-
36

1

Management

1
.1

Consortium management



















1
.2

Meetings



















2

Wave modeling

2
.1

Parameterization



















2
.2

Implementation



















2
.3

Forecast
and hindcast simulation
s



















3

Sea ice modeling

3
.1

Parameterization



















3
.2

Implementation



















3
.3

Forecast
and hindcast simulation
s



















4

Satellite remote sensing of waves
-
in
-
ice

4
.1

Radar backscatter analysis for ice type
recognition



















4
.2

Ice type recognition from SAR and
optical images



















4
.3

Wa
ves
-
in
-
ice retrieval methodology



















4
.4

Co
llocated SAR, optical and CryoSat
data



















4
.5

Wa
ves
-
in
-
ice evolution relative to sea ice
type



















5

Inte
gration and validation

5
.
1

Cross
-
validation of remote sensing
products



















5
.
2

Validation of the wave model



















5
.
3

Validation of the sea ice model



















5
.
4

Integration



















6

Demonstration to users

6
.1

Demonstration
to mar
it
i
m
e transport



















6
.
2

De
monstration of benefits
in

m
et
-
ocean

services



















7

Dissemination and exploitation

7
.1

Public
w
eb
s
ite



















7
.2

Data distribution



















7
.
3

Service sustainability






















Demonstration period




Table 1.3c Deliverables

No.

Deliverable name

WP

Nature*

Dissemination level†

Delivery date

D
1
.1

Progress reports

1

R

PP

Quarterly

D
1
.
2

Project internal website


W

PP

3

D
1
.
3

Final
Activity
report


R

PU

36

D
2
.1

First wave model results

2

D

PP

12

D
2
.2

Upgraded wave model


C

PU

24

D
2
.3

Forecast and h
indcast simulation
s


D

PU

32

D
3
.1

First sea ice model results

3

D

PP

12

D
3
.2

Upgraded sea ice model


C

PU

24

D
3
.2

Forecast
and h
indcast simulation
s


D

PU

32

D
4
.1

Wave and ice type retrieval methodology

4

R, D

PU

12

D
4
.2

Wave attenuation versus ice type and thickness


R, D

PU

24

D
4
.3

Demonstration of wave product in the MIZ


R, D

PU

32

D
5
.1

Validation reports

5

R

RE

24

D
5
.2

Upgraded
NavTracker

software


R

PP

24

D
6
.1

User feedback
report

6

R

RE

24

D
6
.2

Second user feedback report


R

RE

33

D
6
.3

MetOcean
product enhancement report


R

PU

36

D
7
.1

Public web site

7

W

PU

3

D
7
.2

Dissemination plan


O

PU

3

D
7
.3

Scientific and

promotional material


O

PU

12

D
7
.4

Short movie for users in maritime transport


O

PU

30

D
7
.5

Business plan for beyond project duration


R

RE

36

* R: Report, D: Data or Database, C: Source code, W: Web site, O: Other.

† PP: Between partners, PU:
Public, RE: Restricted or confidential.

Table 1.3d Milestones

No.

Milestone name

Work

packages

involved

Expected
date

Means of verification

M1

SWARP Kick off

All

Mo 1

All partners and project
officers attend

M2

First version of wave and sea ice models

WP
5

WP
2
, WP
3

Mo 12


M
3

Validation completed

Data flows integrated to
NavTracker


WP
4
, WP
5

Mo 24

Data reports

and software
documentation

M
4

Service launch

WP
5
,
WP
6

Mo 30

Users attendance to the
launch

M
5

Users uptake

Final report

Final project meeting

All

Mo 36

All partners and project
officers attend


Table 1.3e Workpackage description

Workpackage number

1

Start date or starting event:

Mo 00

WP Package title

Management

Activity Type

MGMT

Participant number

NERSC









Person
-
months per
participant:

4










Objectives

To maintain efficient collaboration between partners, to maintain efficient communication with the
Commission.


Description of Work

Since the distance between partners is large, web
-
based technologies will be used
extensively to verify the
progress and assemble the reports. Quarterly progress reports will be collected and shared between partners.



Conduct financial and administrative management.



Communicate with the EU’s scientific officer.



Lead the project
Scientific Advisory
Committee.



Carry out overall planning of scheduled project meetings, workshops and teleconferences.



Coordinate the preparation of consortium agreement.



Make minutes from the scheduled project meetings and workshops.



Check the overall
quality of deliverables in cooperation with the Quality Controller.



Update participative project web

pages (Drupal or Google Docs based) for general project information and
internal document/data exchange.



Manage IPR, legal, gender equality issues.



Impleme
nt contract amendments if needed.



Compile, produce and distribute the Period Activity and Management reports.



Prepare the Final
Activity
report.



Deliverables


D
1
.1 Progress reports (Quarterly
,

NERSC)

D
1
.2

Project internal website (Mo 3
,

NERSC)

D
1
.3

Fin
al
Activity
report (Mo 36
,

NERSC)





Work

package number

2

Start date or starting event:

Mo 00

WP Package title

Wave modeling

Activity Type

RTD

Participant number

Ifremer

UO

NERSC

ISMER






Person
-
months per participant:

26

-

6

-







Objectives

Develop, implement, and test parameterizations for waves
-
in
-
ice propagation in a wave model.



Description of Work

Task
2
.1 Parameterization

(Ifremer,
UO
)

Develop wave damping

parameterizations based on previous experience on wave scattering by bottom
topography, on air
-
sea friction for swell attenuation, and on wave
-
ice small
-
scale modeling.
Computationally efficient Monte Carlo averaging methods will be designed to scale

up the waves
-
in
-
ice
propagation model from UO
(
a look
-
up table as
from
Bennetts and Squire, 2012)
to the typical resolution of
a realistic wave model.
A dedicated
WAVEWATCH III
numerical model of the Arctic will be imbedded in
the Ifremer multigrid forecas
ting system developed for the IOWAGA (ERC funded) and WAVE
-
DB
projects (funded by U.S. ONR
-
NOPP).

This is a spectral wave model that solves the energy balance equation, and today's full absorption of energy
for ice cover larger than 70% will be replaced by

a dissipation and scattering source term adjusted to the
deterministic wave
-
ice interaction model over the same academic conditions. These source terms will be
later calibrated with remotely sense
d

data. The Arctic WAVEWATCH model grid will use a curvili
near
grid system, already implemented in WAVEWATCH, while still retaining a reference direction to the North
Pole. The effect of wave reflection at shorelines and ice shelves will be added in the curvilinear grid.

Task
2
.2 Implementation

(Ifremer
, NERSC
)

P
erform hindcast simulations of the sea state for particular events (experiments and situations with altimeter
or SAR measurements of waves through ice tongues) to evaluate and refine parameterizations. A first series
of model
results will be produced after

12

months in order to provide a first set of forcing fields for ice
modelling (WP
3
).
A second parameterization and model development iteration will be performed using
inputs of ice types from a sea ice model (WP
3
) and from satellite data (WP
4
).

NERSC will

assist in initial
implementation using their experience of large
-
scale waves
-
in
-
ice modelling.

Task
2
.3
Forecast
and hindcast simulation
s

(Ifremer)

Set up both forecast and
hindcast simulation
s, one for each developmental step
.

During the second step,
inc
lude ice representation from satellite data (WP
4
) and from a sea ice model (WP
3
).

The validation is
carried out in WP
5

starting from the first model results.

Output to the users in WP
5


-

Variables: Significant wave heights, peak period

-

Domain: Global at resolution of 0.5 deg. Downscaling at higher resolution.

-

Hindcast period: 2009
-
2012, Real
-
time forecast starting at Mo 32.


Deliverables


D
2
.1 First wave model results
covering the MIZ
(Mo 12,
data
, Ifremer, PP)

D
2
.2 Upgraded model (Mo 24,
source code
, Ifremer, PU
)

D
2
.3
Forecast and h
indcast s
imulations for demonstration. (
Mo 32,
data
, Ifremer, PU)





Workpackage number

3

Start date or starting event:

Mo 00

WP Package title

Sea ice modelling

Activity Type

RTD

Participant number

NERC

ISMER

NER
S
C

UO






Person
-
months per participant:

19.3

-

18

-







Objectives

The overall objective is to introduce parameterizations for wave related processes for a better representation
of the marginal ice zone by sea ice
models. Particularly, we will develop, test and validate
parameterizations for waves
-
in
-
ice propagation and attenuation, wave
-
induced floe breaking, and ice
formation in wavy conditions. These parameterizations will be implemented in a sea ice
-
ocean couple
d
operational modelling system.


Description of Work

Task
3
.1 Parameterizations

(
NERC
,
NERSC,
ISMER
, UO
)

This task will proceed in two iterative steps in order to improve the parameterization based on intermediate
hindcast simulation results. Wave attenuation induced by the scattering at floe edges requires some
knowledge of the floe size distribution within
a grid cell. Wave attenuation coefficients calculated by
scattering models will be introduced and coupled to a parameterization for floe breaking. Tests and
sensitivity experiments will be performed using one
-
dimensional numerical settings in parallel to
i
mplementation in
the two
sea ice
-
ocean coupled model
s NERSC
-
HYCOM and NEMO
. Wave attenuation
coefficients taking viscous losses into account will be introduced. Model results obtained with and without
viscous processes will be compared.

Two commonly used s
chemes for ocean vertical mixing, the Turbulent
Kinetic Energy (TKE) scheme and Generic Length Scale (GLS) scheme will be optimized for ice
covered
areas using information on the wave field. The optimization
will be
achieved by minimizing the mismatch
between model and observational data, specifically satellite data on ice concentration
,
drift
,
thickness and
floe size distribution

and ocean stratification in the MIZ
.

Task
3
.2 Implementation

(NERSC,

NERC
,

ISMER
)

Implement the parameterizations in the 2D
numerical framework of a sea ice

model. The

first

model chosen
is the
NERSC
-
HYCOM model.
NERC will also implement the parameterizations of Task 2.1 in their sea
ice models that are coupled to the NEMO ocean model. Both groups will t
est the parameterizations and
optimize the code efficiency
.

Task
3
.3
Forecast and
hindcast simulation
s

(NERSC
,
NERC
)

Both real time forecast and h
indcast simulations will be performed for particular events where there are
satellite observations of waves
-
in
-
ice and their effects in order to evaluate and refine parameterizations. A
first
hindcast simulation will be performed to evaluate waves
-
in
-
ice parameterization
, using nesting
techniques to obtain high resolution in areas of interest (the Barents and Kara

Seas)
. Sea ice properties
including the floe size distribution
and the ice type
will be provided to WP
2

at Mo 12
.
A second
hindcast
simulation
will be done with input from an improved wave model (WP
2
) and test new parameterizations
.

The validation is taki
ng place in WP
5
, starting with the first model results.

Output
to the users in WP
5

will be

-

Variables: Sea ice concentrations, sea ice thickness, sea ice type, maximum floe size

(new)
. Ocean
currents, ocean temperature and salinity.

-

Domain: Whole Arctic
at 12 km resolution (TOPAZ), Barents and Kara Sea at 5 km resolution
,
Antarctic at 0.5 deg resolution
.

-

Hindcast period: 2009
-
2012. Real time forecasts starting at Mo 32.


Deliverables

D
3
.1 First
sea ice

model results
including floe size information
(Mo 12,
data
, NERSC, ISMER, PP)

D
3
.1 Upgraded sea ice model (Mo
24
,
source code
, NERSC,
NERC,
PU)

D
3
.2

Forecast and h
indcast simulations (Mo 32,
database
, NERSC,

NERC,

PU
)



Workpackage number

4

Start date or starting event:

Mo 00

WP Package title

Satellite remote sensing of waves
-
in
-
ice

Activity Type

RTD

Participant number

Ifremer

NIERSC

NERSC


ODL






Person
-
months per participant:

6

18

4

18







Objectives

To develop and implement remote sensing methods for observation and model
validation of waves in the
MIZ.


Description of Work


Task
4
.1
Ice type recognition from coarse resolution scatterometry

(Ifremer)

Radar backscatter from ERS, Quikscat and ASCAT data will be collected, collocated and analyzed for
selected test areas of
the MIZ in Arctic and Antarctica. The higher resolution of SAR data compared to
scatterometer data will be used to interpret the scatterometer backscatter data in the MIZ. The relationship
between scatterometer data at a resolution of 12 km and subgrid SAR

information will be used to develop
algorithms for classification of sea ice in scatterometer data. Scatterometer data can provide complete
coverage of MIZ in both hemisphere twice per day and will thus serve as coarse resolution input for wave
models (W
P
2
).


Task
4
.2

Ice type recognition from high resolution SAR and optical images (NIERSC)

This task will analyse the SAR data from ENVISAT and other SAR satellites (e.g. RSAT
-
2, TerraSAR
-
X)
used in task
s

4
.3,
4
.4 and
4
.5 to classify the MIZ in selected test areas where also optical images will be
used. The output of this task will be an algorithm for monitoring the MIZ in SAR images from Sentinel
-
1.
The
MIZ has

a combination of open water and various stages of new and f
irst
-
year ice. The backscatter
values of sea ice depend on
ice types and surface properties such as sea ice salinity, floes size and surface
roughness
.
Several studies have been conducted to determine
sigma
-
0

for different ice types and SAR
configurations,

where

in situ sea ice and snow measurements
are compared with nearly coincident

satellite
SAR data

(
e.g. Sandven et al.

1999).
SAR
sea ice backscatter
studies
will be continued
with
a
focus on the
MIZ, using C
-
band alternating polarization data from ENVIS
AT and other SAR satellites.
The
focus will be
to develop and implement two algorithms for the determination of the zone of broken
-
up floes in the MIZ,
which will be used to validate the floe size distribution given by sea ice models developed in WP
3
.
These
algorithms will use neural network and Bayesian probability methods.


Task
4
.3 Waves
-
in
-
ice retrieval methodology review and implementation (
ODL
)

This task

will select wave in ice imagettes from ERS1
-
2 satellite using data from IFREMER processing
fac
ilities and ENVISAT images from the ESA archive over the Arctic and Antarctic MIZ. The first priority
will be to use SAR data that are contemporary and collocated with optical images (ice
-
water discrimination)
or altimeter data (laser or radar data for ice

thickness). The data will be analysed for wave properties in
relation to ice type and ice thickness. An inversion methodology for wave spectrum retrieval from SAR
images will be implemented. Validation and calibration of the inversion modulation transfer
functions will
be based on wave spectra comparisons in open water and sea ice. Furthermore, observation of waves with
different azimuth directions will be analyzed and compared with in
-
situ wave spectra.


Task
4
.4
Acquisition and analysis of collocated SA
R, optical and CryoSat altimeter data
(NERSC)

The task will focus on acquisition and analysis of collocated SAR, optical and CryoSat altimeter data during
the planned field experiments. The SAR and optical high
-
resolution images will be used for characteri
sation
of the ice condition and wave propagation in the experiment area. The SAR images will be obtained from
RADARSAT
-
2
fine resolution mode

with resolution better than 10 m and TerraSAR
-
X StripMap Mode
with resolution of 3 m. CryoSat freeboard profiles,
with spatial resolution of
3
00 m in azimuth direction
will be acquired and analysed for ice thickness retrieval.
The

satellite data will be used for validation of the
models in WP
5
.


Task
4
.5 Analysis of observed waves
-
in
-
ice evolution relative to sea ice
type (
ODL
)

This
task
will

analyse the retrieved parameters of wave in ice using
the
methodology developed in
4
.3 in
order to quantify attenuation, dispersion and refraction of waves in the MIZ. Furthermore, the observed
wave evolution properties will be related to propagation distance into the MIZ and to ice properties along
the propagation path. Attenuation rate
of wave energy propagating in the MIZ will be derived as
a
function
of ice type and ice thickness.



Deliverables


D
4
.1 Report on wave and ice type retrieval methodology in the MIZ (Mo 12,
report,
ODL
, PU
)

D
4
.2 Report on observed wave attenuation versus ice type and thickness (Mo 24,
report,
ODL
, PU)

D
4
.3 Demonstration wave product in the MIZ from SAR and scatterometer ice type products (Mo 32,
database
,
ODL
, PU)





Workpackage number

5

Start date or
starting event:

Mo
12

WP Package title

Integration and Validation

Activity Type

RTD

Participant number

NERSC

NAVTOR

NIERSC


NERC

Ifremer

ISMER

ODL

Person
-
months per participant:

1
5

20

12

10

8

-



Objectives

Validate
the waves
-
in
-
ice and sea ice
models from WP
2

and WP
3

using satellite and in situ data
.

Integrate these products into
NavTracker
.



Description of Work


Task
5
.
1

Validation of remote sensing products (
Ifremer
,
NERSC
,
NERC
,ODL
)

Input
: Satellite data from WP
4
,
ice charts
from
OSI
-
SAF
,
in
-
situ data from the WIFAR field campaign in
August 2012 (ice thickness, ice type, wave acceleration, weather mast)
, user data (ship ice radar)
.

There are
also further experiments similar to the ones planned for July 2013, involving the deployment of seve
ral
buoys

(up to 10) for 2
-
3 days
.

This is in conjunction with the Office of Naval Research (ONR, USA). If
ONR gives permission

to distribute to all SWARP partners
, this
will
provide wave spectra at several points
within the ice, and
a measurement of the
incident wave spectrum

as well as
observations

of the FSD.

Validate independent satellite data

from WP
4

against each

other

and available in
-
situ data
.

The satellite
products will consist of

1.

Wave
l engths

from SAR i n the MIZ
.

2.

Ice cl assi fi cation from SAR and optical i mages wi th focus on i denti fi cati on of the zone with smal l
fl oes broken up by waves.

3.

Regi onal wind and wave i nformati on from al ti meter, scatterometer and SAR

4.

Regi onal i ce thi ckness i nformati on from the CryoSAT al ti m
eter.


Task
5
.2 Validation of the wave model (Ifremer
,

ODL
)

Input
: Hindcast simulations from WP
2
, satellite data from WP
4
, pre
-
existing in
-
situ data from WIFAR field
experiments
, user data (ship radar)
.

Run the model for the period
of August 2012
when
in
-
situ data from the WIFAR field campaign and
remote
sensing of waves are available and validate the model against data from the same period.

If the in
-
situ data
from the ONR experiment is available, run the model for July 2013 also.

The validation studie
s need to
include periods of different winds, waves and ice conditions.

ODL and Ifremer will also test a novel directional swell sensor
which can retrieve the 2D spectrum at
high
angular
accuracy.

It
will initially be evaluated in ice
-
free conditions befo
re it is deployed on ice
, depending
on opportunities
.
The instrument performance is expected to be good on ice because of the lower noise from
shorter waves.


Task
5
.
3

Validation of the sea ice model (NERSC
,
NERC
,

ISMER
)

Input
: Hindcast simulation
s

from WP
3
, satellite data from WP
4
, in situ
buoy
data

from
IABP
,
ice charts
OSI
-
SAF
.

Run the nested ocean and sea ice model for the same period when
in
-
situ data and

sea
-
ice observations are
available and validate the sea
-
ice model against the in
-
situ data

and satellite data available at the same
period from WP
4

as well as from the OSI
-
SAF. The validation will consider the same period
(s)

as in Task
s

5
.
1 and
5
.2
, and will focus on
the width of the MIZ,
sea ice concentrations,
sea ice type and ice drift
.


Tas
k
5
.
4

Integration

(
NAVTOR,
NERSC
,
Ifremer, ISMER
)

Input
: M
odel
data

from WP
2

and WP
3
.
Agree on data formats and exchange data, this task will start with
the preliminary data at Mo12
so that the format may be adapted to fit the needs of the user before the
start
of operations
.

I
ntegrate
the forecast data
into
NavTracker

NavTracker

& NavPlanner. This includes data harvesting to
Navtor Data Servers, data compression for onboard transmission, and development of new functionality and
presentation in NavTracker
(contingency room/onboard) and NavPlanner (onboard route planning tool) for
optional exchange with the ECDIS bridge system.

Particular attention will be paid to the clarity of the presentation and the geographical and temporal
consistency of the forecast data presented.


Deliverables


D
5
.1
Validation reports (Mo 24,
report
, NERSC, PU
)

D
5
.2 Upgraded software (
NavTracker

&

NavPlanner) including waves
-
in
-
ice and ice forecasts (Mo 24,
report,
NAVTOR, PU)




Workpackage number

6

Start date or starting event:

Mo 31

WP Package title

Demonstration

Activity Type

DEMO

Participant number

NAVTOR

Ifremer

NERSC

NIERSC

NERC

ISMER




Person
-
months per participant:

8

3

6

4

1

-





Objectives

Demonstrate the impact of the proposed service on the operations and capacities of the involved user
communities.


Description of Work


Task
6
.1

Demonstration to

Maritime Transport (NAVTOR,
NERSC
,
Ifremer
, NIERSC
)

N
umerical forecasts of waves and ice will be processed in the Barents Sea
and Kara Sea
accompanied with
special delivery of satellite data in the area
. T
est cases
for validation
will be selected in a period of interest
in the recent

past when high waves reached ice
-
infested waters. The
Fram Strait will be used
in hindcast
mode
since in situ data are available
for all waves
-
ice
-
wind
-
ocean parameters
from the WIFAR field
experiments

in Aug. 2012
.

The
real
-
time forecast
demonstration s
hould be in time for
the opening of the Shtokman field in the Barents
Sea
(2015/2016)
where remote sensing data
will be
available for validation
. These
results
will be presented
to
NAVTOR’s targeted users, to
th攠潩l⁡湤⁧慳⁩湤畳try

i湶潬v敤⁩渠
卨瑯pm慮

慮d

r敬at敤e
畳敲猠
Ein獵s慮ae
捯 灡湩敳 捯 敲i湧 獨i灳p t桥 k潲w敧i慮 C潡ot du慲摳 慮搠t桥 k潲w敧i慮 k慶yF
慴 th攠l慵湣h 潦 the
獥rvi捥
K

q桥hi湦orm慴i潮o灲敳敮t敤⁷ill⁩n捬u摥d



The
product quality information
from WP
5


-

Evidence of stable and reliable data d
elivery (timeliness and availability of service)

-

A demonstration of web
-
GIS
data service
s

from WP
7
.

The users


f敥d扡bk will b攠捯ll
散t敤ed畲i湧 t桥 畳ur w潲k獨潰Ⱐ慮a t桥hr r敱eir敭敮e猠will 扥b捯 l散t敤 in
t敲m猠 潦 摡t愠 煵qlityⰠ 摡d愠 form慴ti湧Ⱐ tim敬in
敳e
I

畳uful湥獳 慮搠 灲e獥ntatio渮

q桥h fe敤扡捫 will be
捯 l散t敤e

煵q獴i潮湡ir敳

慮搠a敬e灨潮攠int敲vi敷s


A
湯n桥r

摥d潮獴r慴i潮o
will 扥bt慲g整敤e at
th攠trai湩ng 潦
y潵湧 潦fi捥r猠t漠famili慲iz攠t桥h t漠th攠湥w
f潲e捡st
a湤nn慶ig慴i潮o
獥rvi捥s

k䕒千 will
畳攠潮o潩湧 捯 l慢aratio湳nwith
th攠k潲w敧i慮 C潡ot d畡rds
Ehy獴v敲k整F
K

f卍䕒 will
i湴敲慣t
wit栠 t桥h 卣桯潬 f潲 潦fi捥r猠 E慴 f湳nitut 䵡ritim攠 摵d nu
é

I

in
oim潵獫iⰠ
http://www.imq.qc.ca/
)

and with FedNav/
Enfotec, in Montréal, Canada
.

NIERSC will use
historical links to the Murmansk Shipping Company.


Task
6
.2

Demonstration

of Benefits in Met/Ocean Services (NERSC,
Ifremer, NIERSC,
NERC
,
ISMER
)

Both public and private met/ocean service providers interested in using the SWARP data will be
approached
to
evaluate the usefulness of the service and their complementarities to existing services.

Examples of such
downstream
services are Search and
Rescue

operations
and
oil spills contingency

in ice
.

The demonstration will
inform
stakeholders from the institutional, research, business and governmental
sectors

including
the Ambassador and Special Advisor for Polar Affairs at the Norwegian Ministry of
Forei
gn Affairs, through NERSC’s participation to the national BarentsWatch portal