Toward integrated ecosystem and biogeochemical modelling in the GODAE framework

1
Toward integrated ecosystem and biogeochemical
modelling in the GODAE framework
Paris, June 12 2007
P. Brasseur, CNRS-LEGI, Grenoble, The French
Alps and many other colleagues (Mercator, CNRS,
CNES Ifremer, IRD, Météo-France, SHOM)

• OUTLINE
• The European context MERSEA Integrated Project
  (FP6)
• The MERCATOR systems for Global and North
  Atlantic oceans
• RD issues and strategy for integrated ecosystem
  modelling
• The future EuroARGO and MyOcean (FP7)

2
The GODAE IMBER initiative (Ecosystem
modelling)

• To facilitate dialogue between those developing
  new ecosystem models and the developers of the
  operational systems.
• To promote mutual understanding of the
  requirements of the two communities.

• Areas of importance
• Ecosystem modelling data assimilation
• - Schemes for assimilation of
  biogeochemical data are under development
• - Current assimilation schemes degrade the
  biogeochemistry
• - Overall high horizontal and vertical
  resolution models for the upper ocean are
• needed.
• - Advanced schemes for a finer vertical
  structure are a key issue for nutrient
• transport.
• Interaction with coastal and shelf seas systems
  
• Support for B-Argo (see friends of Oxygen on
  Argo)
• Reanalysis

First GODAE-IMBER Meeting Paris, France, 12-13
June 2007
Establishment of GODAE-IMBER Working Group
desired to coordinate modelling and
observations
3
Ecosystem response to oceanic variability ?
4

• 1. European framework for GODAE the MERSEA
  Integrated Project (FP6)

5
Development of a European system for operational
monitoring and forecasting of the ocean physics,
biogeochemistry, and ecosystems, on global and
regional scales
6
MERSEA Integrated Project

• RD project funded under 6th FP of the European
  Commission
• Thematic priority SPACE - GMES (Global
  Monitoring for Environment and Security)
• Ocean and Marine Applications
• Four-year project (2004 2008)
• 40 contractors, 16 countries (or Int. Org.)
• Coordination IFREMER (Institut Français de
  Recherche pour lExploitation de la Mer), France

7
MERSEA Global to Regional coverage
8

• 2. The MERCATOR systems for Global and North
  Atlantic oceans

9
Incremental development 3 main steps
PSY3
PSY2
PSY1
10
Mercator systems running today and short term
upgrades

• Global ¼
• Before April 2007 Global ¼ with optimal
  interpolation (lifting and lowering method) of
  altimetry alone, based on OPA8,  if-type  ice,
  daily forcing (fluxes), weekly operations
• Since April 2007 update of the ¼ global
  OPA/NEMO, LIM-2 ice, SEEK assimilation (3D EOFs)
  of altimetry, T/S profiles, SST, daily forcing
  (BULK formulae, weekly operations
• Both systems operated until end of september (end
  of Mersea TOP2)
• North Atlantic and Mediterranean HR
• Today PAM grid 5 to 7 km North Atlantic (gt9N)
  and 1/16for the Mediterranean, OPA8, optimal
  interpolation using multivariate statistics
  (including 1D vertical) of altimetry, T/S
  profiles and SST, daily forcing (fluxes), weekly
  operations
• Summer 2007 update using NATL12 OPA/NEMO 1/12
  ORCA grid, SEEK assimilation of altimetry, T/S
  profiles and SST, daily forcing using BULK
  formulae, weekly operations

11
Towards daily operations
Mercator systems medium term upgrades

• High temporal resolution (lt6h) forcing under
  investigation
• Operational daily scenario under investigation
• Forcing updates and assimilation
• April 2008 start of a demo system on the North
  Atlantic
• August 2008 operational daily service

Towards global ¼ reanalyses of the physical ocean

• ORCA025-LIM SAM2 (SEEK)
• 2007 first stream 2002-2007 (Mercator years)
• 2008 second stream 1993-2007 (modern altimetry
  years)
• 2009 third stream 40 years (ERA40)

12
Coupled models NEMO OPA9
PISCES O(100) parameters (Aumont et al., 2004)
LOBSTER (Lévy et al., 2003)
13
Implementation
Application
RD
WP4 (O. Aumont) Interface to regional sub-system
WP1 (M.Levy) Improvement of process resolution
PSY3-PISCES WP3 (C. Moulin) Extension to
global ocean PSY4-PISCES
WP2 (P. Brasseur) Model-data integration
WP5 (P. Lehodey) Extension towards marine
resources
WP6 Evaluation strategy (I. Allen)
 Green Mercator 
14
Toward a bio-component in MERCATOR
PISCES 1 real time
june 2007
june 2008
PISCES 1/4
NATL12
april 2009
PISCES 1/4 real time
ORCA12
15

• 3. RD ISSUES and STRATEGY for integrated
  ecosystem modelling

16
Data assimilation for coupled physical-biogeochem
ical modellingmultiples sources of uncertainty
!

• Hydrodynamics

Relevant to GODAE!

• Biology

17
MERSEA RD OBJECTIVES

• OPERATIONAL
• O.1 Develop the prototype of a coupled
  physical-biological assimilative system, which
  will be operated in delayed mode during selected
  periods (1998) to routinly estimate and monitor
  biogeochemical variables
• O.2 Transition the tool (or pieces of it) for
  operational demonstration on the North Atlantic,
  in connection with other MERSEA activities

• SCIENTIFIC
• S.1 Quantitatively improve the representation
  and space-time variability of marine ecosystems
  in ocean basins, by coupling with multi-data
  assimilative eddy-resolving circulation models,
  focusing on key coupling mechanisms (mixed layer
  dynamics, eddy activity, vertical advection,
  etc.)
• S.2 Investigate the scientific value of ocean
  colour to improve the representation of primary
  production, and develop new approaches to invert
  this information in models

18
System available at MERSEA start Integration
strategy

• QUESTIONS
• How does the assimilation of multivariate data
  sets affect the coupling and ecosystem response ?
• How should the assimilation platform be modified
  to improve the coupling
• circulation model and forcings ?
• ecosystem model ?
• assimilation scheme ?

19
Platform V0 surface Chl_a (mg m-3) , March 1998
SLA SST assimilation
No DA

• Free model simulation over-estimates the
  observed data
• The assimilation intensifies the biological
  response

SeaWiFS
20
Vertical diffusivity at 30W
No assimilation
SLA SST assimilation
21
Nutrient supply (integrated in euphotic zone)
22

• NEW RD ASSIMILATIVE PLATFORM
• Ocean circulation model NEMO-NATL4
• Biogeochemical model LOBSTER
• Assimilation engine SEEK-IAU

23
NEMO circulation model(Barnier et al. - DRAKKAR)
North Atlantic configuration Domain 20S-80N,
98W-23E Resolution 1/4 - 21 km in the Gulf
Stream 45 levels, 6m to 250 m Free
surface formulation, partial steps Surface
fluxes ERA40 (bulk formulations) SSS relaxation
to Levitus climatology Experimental
set-up Spin-up period interannual 1984-1996
Hindcast period 1996-1998 Assimilation scheme
SEEK filter, local error sub-space (EOFs),
incremental updates Assimilation data
along-track T/P ERS altimetry, AVHRR SST and
climatological SSS
24
LOBSTER ecosystem model (Levy et al., 2002)

• Coupled on-line to circulation model
• Bio parameters tuned for North Atlantic basin

25
The SEEK assimilation cycle
26
Modified SEEK analysis Vertical stratification
(1D tests)
Unconstrained analysis
Constrained analysis
Forecast
Observations
27
1998 EXPERIMENTS
Exp. Name FREE NI_A PH_A PN_A
Assimilation of physical data No No Yes Yes
Assimilation of nitrate fields No Yes No Yes
Control Vector None NO3 U, V, T, S U, V, T, S, NO3
Data assimilated None NO3 climatology SST, SSH T and S climatology SST, SSH T, S and NO3 climatology
28
1998 EXPERIMENTS
Exp. Name FREE NI_A PH_A PN_A
Assimilation of physical data No No Yes Yes
Assimilation of nitrate fields No Yes No Yes
Control Vector None NO3 U, V, T, S U, V, T, S, NO3
Data assimilated None NO3 climatology SST, SSH T and S climatology SST, SSH T, S and NO3 climatology
NO3 clim backgroud term
29
(No Transcript)
30

• FREE
• PH_A
• NI_A
• PN_A

Misfit to CLIM nitrate
31
(No Transcript)
32
(No Transcript)
33
(No Transcript)
34
1998 Phyto-Zoo grazing flux (mmolN.m-2)
FREE
NI_A
PH_A
PN_A
35

• 4. Future plans in Europe EuroARGO and MyOcean
  (FP7)

36
EURO-ARGO

• Context ESFRI (European Strategy Forumon
  Research Infrastructures) (European Commission).
  Roadmap for new large research infrastructures of
  pan-European interest.
• Euro-Argo selected (35 projects 7 environment
  sciences)
• Proposal Europe establishes an infrastructure
  for ¼, i.e. 800 floats in operation
• Requirement 250 floats per year including
  regional enhancements (Nordic, Mediterranean and
  Black seas)

37
EuroARGO - Preparatory Phase Proposal

• Main expected outcomes
• Agreement for long term (10-20 years) operation
  of Euro-Argo (financial, legal, governance,
  organisation, technical). Member States
  (ministerial level)
• Agreement with EC (GMES, GEO, DG Research) for
  additional long term EC funding
• Main technical and organizational issues to be
  solved
• Links with international structure

38
Optical sensors implementation on PROVOR (PROVBIO
II) (Claustre and DOrtenzio, LOV)
PROVBIO II
39
Future developments and perspectives
40
GMES Marine Core Serviceand European operational
oceanography
00
01
02
03
04
05
06
07
08
09
10
11
12
13
...
Operational
Implementation
Initial
GMES Phases
Demonstrate the European maturity of oceanography
Build the GMES marine system, integrating the
core capacities (MERSEA)
Run the GMES marine core service, on an
operational basis

• Challenge (1) run the European core service on
  an operational basis
• Challenge (2) link definitively with the
  European and member states main services and
  applications
• Challenge (3) organise an a sustainable basis
  the link between this operational European
  service and the existing research networks

41
MY OCEAN a 3-year project to set up and
operate the GMES Marine Core Service
The down stream Cut off
The upstream cut-off
downstream to our service ... is done (duty), or
will be better done (skill) by a specialized
agency, a European agency or a national center
usually already in place Example COASTAL
SYSTEMS
upstream to our service ... is done (duty) by an
observation agency or center (raw data) Example
Eumetsat SAF or the ESA PAC
Data, Model European added-value
42
Modelling and Forecasting centers and regions

• 1. Global
• 2. Arctic
• 3. Baltic
• 4. NWS
• 5. IBI
• 6. Med Sea
• 7. Black Sea

Arctic GOOS
2
NOOS
4
3
BOOS
1
GOOS/ GODAE
1
7
1
6
Black Sea GOOS
5
6
IBI-ROOS
MOON MedGOOS
43

• 5. Conclusions

44

• The assimilation of physical data can improve the
  representation of ecosystem variability (in space
  and time) on many aspects (upwellings, mesoscale,
  high-latitudes, ), BUT existing (operational)
  methods have not been designed in that
  perspective
• Need to constrain not only the physics but also
  the components of the biological state vector
• The tools to assimilate, e.g. nitrate data O.K.,
  but we need reliable in situ data ! Impact
  studies should be performed to recommend
  strategies for in situ bio measurements which
  variables, sampling strategies ?
• Metrics to evaluate the benefit ? Toward
  BIO-MERSEA metrics ? Too early ?
• Useful ocean information and required accuracy
  for computing diagnostic quantities which cannot
  be measured directy (grazing, regenerated
  production, pCO2, ) ?
• Ocean colour should be assimilated in the future
  observability ? Assimilation of OC products or
  direct assimilation of SSR ?