Coastal TT

1
Coastal TT

• Pierre De Mey, LEGOS
• Villy Kourafalou, U. Miami

2
Coastal ocean specifics

• Focus shelf break exchanges, shelf dynamics,
  coastal current associated (sub-)mesoscale
• Influence of coastal ocean processes felt far
  beyond shelf break, overlaps interacts with
  open ocean dynamics
• Continuous spectrum of LF and HF processes with
  specific physics free surface, tides, 3D
  processes, etc.

3
Outline

• Previous activities (2006-2008)
• Where do we go from here?
• Towards ToRs
• ICODAE Core Science examples and added value for
  GOV
• ICODAE first steps

4
Previous activities (2006-2008) -- 1

• Explicit interface with Coastal and Shelf Seas
  came in late in GODAE schedule
• Assess illustrate value of GODAE results for
  coastal and shelf seas models and forecasting
  systems
• Build a community
• Coastal and Shelf Seas Working Group
• Permit a first census of national,
  operational-track CSS activities using GODAE
  products
• Members 1 US, 1 CAN, 2 EU, 1 JP, 1 CN, 1 AUS
• 40 systems on web site
• Touch wider community
• Community forum events
• Coastal session at the 2006 Beijing GODAE
  Symposium Ocean Data Assimilation and Prediction
  in Asia-Oceania (16 abstracts)
• 2007 Liverpool Workshop "Assessing the value of
  GODAE products in coastal and shelf seas (36
  abstracts)
• Special session at 2008 Fall AGU meeting, San
  Francisco Advances in the Prediction
  Capabilities of Interdisciplinary Nested Models
  in Coastal and Shelf Seas (1 oral 2 poster
  sessions, 40 abstracts)
• Poster sessions at 2008 GODAE Final Symposium in
  Nice.

5
Previous activities (2006-2008) -- 2

• Publications
• Web pages on GODAE site www.godae.org/CSSWG.html
• White Paper (2nd ed., Oct 2007) Towards the
  assessment and demonstration of the value of
  GODAE results for coastal and shelf seas models
  and forecasting systems
• An attempt at a state-of-the-art summary (as of
  Oct. 2007)
• Available on web site
• OcDyn special issue (out, Feb 2009) Assessing
  the value of GODAE products in coastal and shelf
  seas
• 8 papers from Liverpool workshop results
• Good set of results on downscaling from GODAE
  products
• Final Symposium summary paper in OcMag (accepted,
  May 2009) Applications in coastal modelling and
  forecasting
• Added value of downscaling from GODAE systems,
  with examples not shown in OcDyn
• Elements influencing predictability w/examples,
  initialization, data assimilation

6
GODAE CSS Community
P19
P18
P20
P21-22
P23-24-25-26
P29-37
P27
P6 P40
P16
P28-31-38
P12
P17
P7
P30-39
P32-33-34-35-39
P9-10-11
P8
P10-13-14-15
P4
P36
P31
P5
P3
P2
P1

• 40 projects/systems/groups
• Using b.c.s from GODAE core systems
• Varied in objectives and methods
• Geographically clustered
• Africa 1
• Australia 2
• China Seas 2
• Japan Seas 4
• Indian Ocean 1

• North Pacific 1
• North America West Coast 4
• Gulf of Mexico 4
• North America East Coast 2
• Arctic and Nordic Seas 3
• Norwegian, Baltic, and North Seas 7
• Northeast Atlantic 7
• Mediterranean 5

7
GODAE CSS Community
P19
P18
P20
P21-22
P23-24-25-26
P29-37
P27
P6 P40
P16
P28-31-38
P12
P17
P7
P30-39
P32-33-34-35-39
P9-10-11
P8
P10-13-14-15
P4
P36
P31
P5
P3
P2
P1

• 40 projects/systems/groups
• Using b.c.s from GODAE core systems
• Varied in objectives and methods
• Geographically clustered
• Africa 1
• Australia 2
• China Seas 2
• Japan Seas 3
• Indian Ocean 1

• North Pacific 1
• North America West Coast 4
• Gulf of Mexico 4
• North America East Coast 2
• Arctic and Nordic Seas 3
• Norwegian, Baltic, and North Seas 7
• Northeast Atlantic 7
• Mediterranean 5

8
CSS Community sharing experience
OcDynYYYYY(Y)Y

P40 Kyushu U./Fisheries Institute
Naoki
Hirose JP Y
9
CSSWG White Paper contents

• 1 Rationale
• 2 Coastal and shelf seas downscaling within GODAE
• Uses of coastal models
• Links between large-scale and coastal models
• Ongoing/planned CSS downscaling projects linked
  to GODAE
• 3 Issues and requirements related to downscaling
• Critical issues and requirements
• Open issues
• 4 Proposed approach
• Coastal workshop(s), links with international
  activities, etc.
• Appendices
• CSSWG and contributors to PP
• Details of CSS projects
• Details of large-scale GODAE data producers
  models, products, data serving strategy
• The MERSEA and ECOOP European projects

- Access to adequate large-scale GODAE
simulations - Jump in ocean physics and in
forcing functions - Initialization - Tides and
barotropic dynamics - How do we measure goodness
and benefit in CSS models? - Data assimilation
in CSS models - Two-way coupling - Unstructured
grid modelling
10
OcDyn Special CSSWG issue (February 2009)

• Special Issue editors De Mey, Proctor,
  Halliwell, Powell
• De Mey Proctor, Editorial
• Kourafalou et al., Evaluation of GODAE boundary
  condition effects in South Florida nested
  simulations
• Halliwell et al., Impact of GODAE products on
  nested HYCOM simulations of the West Florida
  Shelf
• Powell Moore, Estimating the Analysis Error
  from 4DVAR Data Assimilation of GODAE products
• Proctor et al., The influence of initial and
  open boundary conditions on POLCOMS in the North
  East Atlantic
• Alvera-Azcarate et al., A nested model of the
  Cariaco Basin (Venezuela) description of the
  basin's interior hydrography and interactions
  with the open ocean
• Counillon Bertino, High resolution ensemble
  forecasting for the Gulf of Mexico eddies and
  fronts
• De Mey et al., Assessment of observational
  networks with the Representer Matrix Spectra
  Method
• Herzfeld, Improving stability of regional
  numerical ocean models

11
Where do we go from here?
12
Prospective from last IGST (Washington, D.C.,
June 2008)

• Could envisage transition from present situation
  two types of actions
• Keep/develop forums to discuss cross-cutting
  science issues (some old, some new) and promote
  networking btw. CSS groups systems
• Downscaling-related (modelling, observing system
  e.g. SWOT, usability of large-scale products, HR
  bathymetry)
• Interdisciplinary (circulation, tides, probably
  ecosystem)
• WP issues DA (e.g. w/tides), ensembles, 2-way
  nesting, upscaling, unstructured grids,
  added-value eddy-resolving, atmospheric forcing
  issues, HF, tidal mixing,
• Metrics for assessment and intercomparison
• Opportunities of linkinghosting project of IOC
  wg on Marine Modelling (science issues), JCOMM
  expert team (e.g. intercomparison standards of
  CSS systems), MyOcean science board, GCOS
• See CLIVAR panels structuring
• Foster establishment of pilot projects in
  physically-meaningful regional ocean domains
• Help bring focus and weight to science
  discussions
• Define geographically-meaningful metrics and
  intercomparison
• Take final users demand into account and relay to
  supplying groups
• A way to include more systems, while remaining
  manageable
• Contribution of MyOcean MFCs, links w/JCOMM
  regional services?

13
Coastal Ocean and Shelf Seas Task Team (Coastal
TT)

• Defining TT mission work towards the provision
  of a sound scientific basis for sustainable
  multidisciplinary downscaling and forecasting
  activities in the world coastal oceans
• Time frame 4 years (2013)
• Main actions
• Continue co-sponsoring forums (AGU, EGU) to
  discuss cross-cutting science issues
  (downscaling, predictability, interdisciplinary)
• Remain visible to community of coastal modellers
• Promote limited-time Pilot Projects (PPs)
• CSS forecasting largely supported at national
  level
• Operational implementation not always stabilized
  at institute/agency level
• Focus on coastal forecasting science (not on
  coastal forecasting coordination)
• Fewer groups but more tightly integrated than
  former WG
• In several world coastal regions
• Using GOV products
• Integrated in terms of science and objectives -gt
  next slide
• Contribute to the training of young scientists

14
ICODAE - 1

• ICODAE International Coastal Ocean Data
  Assimilation Experiment
• A label for Pilot Projects (or existing projects)
  adhering to the ToRs
• Types of activities (developed in ToRs)
• Conduct core science investigations related to
  the target processes (-gt Villy)
• e.g. submesoscale, shelf dynamics, biology,
  validation,
• Interface with (some) end users and conduct
  application-targeted science investigations
  related to the target applications
• e.g. forecasting, atmosphere interaction, biology
  (upper trophic levels), ...
• Coordinate activities and results within TT,
  relate with GOV operational activities
• All of the above open to coordination, in
  particular tools metrics, downscaling,
  assimilation, downscaling error estimates (if
  available from large-scale systems)
• Neville Smith I think the challenge is to find
  those elements that are shared and would
  encourage this nascent coastal GODAE community to
  work together. I think one element is the
  development of a discipline for evaluating the
  value of models, and the level of predictability
  (growth of errors). If this group could become
  the developer of standards, I think it would be a
  great contribution.
• Plan (inter-regional) intercomparison, array
  design studies
• Feedback onto GOV
• Contribute to training and forums related to
  coastal forecasting
• Summer school on coastal forecasting ca. 2012
• PP representatives members of Coastal TT

15
ICODAE - 2

• Focus regions for PPs (non-limitative list)
  (readiness)
• Gulf of Mexico
• US West Coast
• North-East Atlantic (NWS/IBI/National)
• Mediterranean
• Japan seas
• Other australian/australasian region?
• ?

16
Coastal ocean forecasting applicationsUseful
forecast lead times -- sorted by variable

• Sea Level
• Storm surges -- O(1day)
• Surface Currents (SC), Nearshore Currents (NC),
  Flags
• Maritime safety (SC) -- searchrescue hours
• Oil industry (platform maintenance, oil spills)
  (SC) -- hours to days
• Nuclear industry (radionuclide spills) (SC) --
  days
• Ship routing, sail races (SC) -- days
• Maritime safety (SC,sea ice detection flag) --
  icebergs days to weeks
• Sediment transport (beaches) (NC) -- days to
  weeks
• Temperature (T), Density (DENS) Vorticity
  (VORT)
• Response of surface layer to HF wind/pressure,
  tropical cyclones, hurricanes (T,SC) -- 1day to
  several days
• Coastal upwellings -- days
• Military (sonar prediction) (DENS,VORT) --
  O(1week)
• Biophysical parameters (primary production,
  fisheries, aquaculture, swimming water quality,
  sewage spills) (DENS,VORT) -- days to weeks
  (requires small-scale vorticity currents)

17
Ecological modelling of nutrients and
phytoplankton(PREVIMER, France)
18
Observation-related items open questions

• Coastal TT missions related to in situ observing
  systems?
• GOOS Regional Alliances
• MOON, ROOS (e.g. IBIROOS), OOI, tide gauge
  networks, ...
• Stay aware of regional databases QC procedures
  -- through PPs
• Promote interoperability free access
• How can we provide support/guidelines?
• Coastal TT missions related to space observing
  systems?
• How can we provide support/guidelines to SWOT,
  OST, other space missions of interest for
  coastal oceanography (new or to be sustained)?
• COSPAR in Bremen in 2010? Session A21 -- ...
  space-borne observations, and their applications
  in the Coastal Zone (currents, algal blooms,
  freshwater inflow)

19
Core Science and added value for GOV

• Predictability
• COSS downscaling/forecasting as added value to
  GOV
• (coastal physics, land-sea interactions,
  strait dynamics, wetting/drying, tides,
  local data)
• multi-scale modeling and observations
• (time, space)
• data assimilation issues / array design
• interdisciplinary (particle dynamics,
  biophysical modeling low/high trophic,
  from nutrient transport to fisheries)
• Coastal to offshore interactions
• Shelf break processes, inter-basin exchanges,
  flows through passages, connectivity
• Upscaling / improvement of predictability of
  the outer models (2-way nesting,
  synthetic observations)
• Specific applications (emerging)
• Extreme events (hurricanes)
• Climate change (the current climate models
  cannot properly address impacts on coastal
  sea level etc.)

20
OS52B - Advances in prediction capabilities of
interdisciplinary nested models in coastal and
shelf seas II

• (Topic 1) "Downscaling adding value to
  large-scale products"
• Ocean Prediction via downscaling of large-scale
  ocean models (Hogan et al., invited)
• Downscaling Global Model for Continental Shelf
  off SE Australia (Roughan et al.)
• High-resolution oc/atm modeling of the Florida
  Big Bend region (Morey et al.)
• (Topic 2) "Assimilation in coastal and shelf seas
  systems"
• Stochastic methodologies to assess coastal
  observational networks (Le Hénaff et al.)
• Discussion Enhancing predictability in the
  coastal ocean open topics
• (Topic 3) "Validation of coastal ocean models"
• Interdisciplinary modeling of the California
  Current System (Edwards et al., invited)
• Coastal Observation and a Forecasting System for
  German Bight (Stanev Colijn)
• Impact of GODAE i.c.s on Coastal Response to
  Hurricanes (Halliwell et al.)
• (Topic 4) "Utility of coastal models"
• Biophysical Larval Tracking System (Paris et al.,
  invited)
• Sensitivity of Irish Sea Temperature to Inherent
  Optical Properties (Polton et al.)

21
Assimilation in coastal ocean systems
22
Data-based validation of coastal-ocean systems
Influence of downscaling and DA at South Florida
outer shelf mooring site during a period of
intense Gulf Stream eddy activity (P14, RSMAS)
23
Data-based validation of coastal-ocean systems
24
Added value tides

• 1/12 Northeast Atlantic NEMO configuration
  (target 1/36 in MyOcean GMES project)
• Downscaled from MERCATOR GODAE product

with tides
Modis
No tide
July 2004 tidal mixing fronts impact on SST
(Chanut, MERCATOR)
25
Added value tides
Mouton SHOM) MERCATOR
Observations 13 July 2004
26
Coastal to offshore interactions Example
Eddy activity and fronts in the Gulf of
Mexico
27
Added Value to GODAE products
NA-HYCOM 1/12 deg (GODAE) Daata Assimilative (OI)
28
Implications on large scale predictability
influence of eddies on Gulf Stream meandering
Apr 21 2004 SeaWiFS data
(provided by Chuanmin Hu, USF)
29
Synergy of high resolution observations and high
resolution modeling
MODEL (sub-domain)
Jan 19 2005 00Z
Jan 20 2005 00Z
Jan 21 2005 00Z
WERA data (N.Shay, UM)

• High resolution coastal modeling
• resollve submesoscale features
• provide 3-D fields to analyze surface currents
  from high frequency radar (WERA)

30
Implications on connectivity and
fisheries Relative influence of
physical-biological interactions on larval
dispersal
2
38
12
28
PLD 30 d
27
15
0.2
13

• Ontogenetic vertical migration behavior
• (damselfish Stegastes partitus)
• increased overall recruitment
• changes in dispersal direction
• changes in larval linkages

OVM
Passive
(Paris et al. 2007)
31
Hurricane prediction regional GoM-HYCOM with
HWIND (NOAA-AOML)
G. Halliwell (NOAA)
32
ICODAE Example Partnership for International
Research and Education International Coastal
Data Assimilation Experiment in the Caribbean and
Mediterranean Seas (pre-proposal to NSF)

• Objectives
• combine biophysical in-situ and satellite
  observations, models and data assimilation
  techniques suitable for the dynamical downscaling
  of global oceanographic fields to the
  shelf/coastal regions toward forecasting and the
  design of optimal observing systems
• resolve time and spatial scales that are
  important for the cross-marginal and inter-basin
  exchange of waters, larvae and nutrients that are
  not currently provided by large (global) scale
  products
• embrace an interdisciplinary and international
  approach to research, education and outreach that
  will have a long lasting impact on coastal
  science, management and policy.

RESEARCH GOAL development and skill assessment
of interdisciplinary, data assimilative coastal
and regional numerical models integrated with
coastal observing systems and nested within
global models to achieve dynamic downscaling of
both models and observations from the global to
the coastal scale
EDUCATIONAL GOAL prepare tomorrows scientists
by providing direct research experience and
interdisciplinary training at international
academic and industry science centers of
excellence and industry
33
PIRE-ICODAE US Partners
University of Miami (lead Institute) V.
Kourafalou (lead PI) Co-PIs and Senior Personnel
(A) observations (in-situ) R. Cowen, B. Haus,
N. Shay (B) observations (satellite) H. Graber
(Director, Center for SE Tropical Advanced Remote
Sensing), P. Minnett (C) modeling C. Paris, J.
Olascoaga, A. Reniers (D) data assimilation A.
Srinivasan, A. Griffa (E) education
assessment A. Bessell (Education department),
S. Sponaugle (NSF/Advance SEEDS Scientists and
Engineers Expanding Diversity Success) (F)
applications J. Ault (fisheries), J. McManus
(reef ecology), L. McManus (policy and
management) (G) international collaboration P.
Ortner (Director, CIMAS Coop. Inst. Mar. Atm.
Science, CaraCOOS member) Industry Partner M.
Roffer, Ocean Fishing Forecasting Service Inc.
(ROFFS/Caribbean Mediterranean)   INFORMATION
MANAGEMENT Center for Computational Science (N.
Tsinoremas, Director) PROJECT OVERSIGHT
International Steering Committee, Internal
Steering Committee and External Advisory
Board RECRUITMENT, ORIENTATION AND
OUTREACH RSMAS Communications (B. Gonzalez,
Director) Florida State University (PI E.
Chassignet)coupled ocean-atm. and biophysical
modeling University of South Florida (PI R.
Weisberg) real time observations and
modeling University of the U.S. Virgin Islands
(PI N. Idrisi) Caribbean data and real time
modeling University of Puerto Rico (PI J.
Corredor) Caribbean data and real time
modeling U.S. Institutes/Affiliates NOAA-AOML
(G. Halliwell) Observing System Simulation
Experiments (user) NOAA-Southeast Fisheries (J.
Lamkin) fisheries research (user) MOTE
laboratory (R. Hueter) fisheries research
(user) National Marine Sanctuaries (B. Keller)
ecosystem management (user) NRL-MRY (J.
Cummings) High resolution atmospheric data sets
(provider) NRL-SSC (P. Hogan) Global model
archives and downscaling (provider) Nature
Conservancy (P. Kramer) environmental research
(user)
34
PIRE-ICODAE International Partners
Belgium P1, Univ. of Liege, A. Barth (D) Brazil
P2, U. of Sao Paulo, E. Campos (A,C), P3, U. of
Bahia, R. Silva(C), NSF/PIRE on Amazon France
P4,LEGOS (Laboratoire d'Etudes en Geophysique et
Oceanographie Spatiales), P. de Mey (D) P5,
CNRS-LA (Centre National de la Recherche
Scientifique), C. Estournel (C) P6, SHOM
(Service Hydrographique et Oceanographique de la
Marine), Y. Morel (C) Greece P7, HCMR (Hellenic
Center for Marine Research), K. Nittis, MedGOOS
chair (A, D, F) Italy P8, NURC (NATO Undersea
Research Center), M. Rixen (B,C,D) P9, INGV
(Istituto Nazionale di Geofisica e Vulcanologia),
P. Oddo (C,D) P10,OGS (Oceanografia e Geofisica
Sperimentale), P. Poulain (A) P11, U. of
Bologna, N. Pinardi (C, D) P12, CMCC
(Euro-Mediterranean Centre for Climate Change),
S. Dobricic (C, D) P13, CNR (Consiglio Nazionale
Delle Ricerche), S. Vignudelli, L. Santoleri (B,
C, D) Mexico P14, CINVESTAV (Centro de
Investigación Estudios Avanzados/Instituto
Politécnico Nacional), D. Aldana (A,F) Norway
P15, NERSC (Nansen Environmental and Remote
Sensing Center), L. Bertino (D) Spain P16,
IMEDEA (Instituto Mediterráneo Estudios
Avanzados/U. Balearic Islands), J. Tintore (A,
F) Industry Foreign Affiliate Partner P17, Delft
Hydraulics/ DELTARES, Netherlands (J. Van Gils)
(C, F)   Affilation with International Programs
GODAE/OceanView (GOV), GOOS (Global Ocean
Observing System), GEO/CZCP (Group on Earth
Observations/Coastal Zone Community of Practice),
LOICZ (Land-Ocean Interactions in the Coastal
Zone), HYCOM Consortium for Data Assimilative
Modeling, CaraCOOS (Caribbean Coastal Ocean
Observing System), MOON (Mediterranean
Operational Oceanography Network), European
Commission Education and Training ERASMUS MUNDUS
Program (UM is a member)
35
Enhancing predictability in coastal-ocean
systems

• Strategy wrt. downscaling initialization?
• Dynamical downscaling nesting (1- or 2-way)
  grid refinement approaches
• Estimation-based downscaling use larger-scale
  estimates as bits of information, optionally
  within (weak) constraints (e.g. gravity-wave
  transients)
• Strategy wrt. atmospheric forcing products?
• Forecast range, space/time effective resolution,
  sensitivity to coastal orography
• Coupled ocean/atmosphere coastal effects
• Multiscale modelling
• Which uses for two-way nesting,
  unstructured-grid modelling?
• Data assimilation strategy?
• Complex error subspace non-stationary,
  non-homogeneous, non-Gaussian, open to lateral
  atmospheric error forcing
• Are advanced methods preferable (4D-var, EnKF,
  full SEEK) ?
• Interest of augmented state vector (physical vars
  surface atm others) ?
• Array design
• Rich spectrum of available data types (HF radars,
  tide gauges, coastal observatories, gliders,
  gravity)
• Detect model state errors
• Control model state errors via DA
• Objectively test models (are pdfs compatible?)

36
COSST outlook Y1 steps

• Finish drafting ToRs summer
• 1-2 PPs by end of year would be a good start!
• possibly ICODAE/PIRE (GoM, Med) or other
• Importance of intercomparison
• Contact groups populate TT
• PP representatives (evolving)
• Experts (key issues, invite papers to forums)
• Need national representatives as in CSSWG?
• Coordinate w/other TTs - enhance GOV
  socioeconomic component
• Organize special sessions as forums
• 2010 AGU Ocean Sciences meeting, Portland,
  Oregon, USA -- session proposal pending
• Session proposal for 2011 EGU

37

• Thank you.