1 / 41

Canadian Report GODAE Ocean View Science Team June 2009

Fraser Davidson (1), Hal Ritchie (4,6) Greg Smith(1,4), Andry Ratsimandresy(1), Debbie Anne Power (1) Adam Lundrigan (1), Charles Hannah (2), Frederic Dupont (2), Dan Wright (2), Maud Guaracino (2), Denis Lefaivre (3), Pierre Pellerin(4), Mark Buehner (4),

gaetan
Télécharger la présentation

Canadian Report GODAE Ocean View Science Team June 2009

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Fraser Davidson (1), Hal Ritchie (4,6) Greg Smith(1,4), Andry Ratsimandresy(1), Debbie Anne Power (1) Adam Lundrigan (1), Charles Hannah (2), Frederic Dupont (2), Dan Wright (2), Maud Guaracino (2), Denis Lefaivre (3), Pierre Pellerin(4), Mark Buehner (4), Lt. Darryl Williams (5) Keith Thompson (6) (1) DFO-Northwest Atlantic Fisheries Centre, St. John's, NL, Canada (2) DFO-Bedford Institute of Oceanography, Dartmouth, NS, Canada (3) DFO-Institute Maurice Lamontagne, Mont-Joli, QU, Canada (4) EC-Meteorological Research Division, Dorval, QU, Canada (5) National Defence, METOC, Halifax, NS, Canada (6) Dalhousie University, Halifax, NS, Canada Canadian Report GODAE Ocean View Science Team June 2009

  2. Canadian Operational Network of Coupled Environmental PredicTion SystemsCONCEPTS • MOU among 3 Government Departments • No $ • Provides ability to obtain funding • Short term (projects) • Long term (budgeted development) • Secretariat provided by • Marty Taillefer DFO and Pierre Pellerin

  3. CONCEPTS: CORE PROJECTS Project 1: Core CMC Systems Installation Install common core ocean model configurations at Canadian Meteorological Centre (CMC) for collaborative projects, evaluations and coupling with GEM atmospheric model, coupled data assimilation, examine potential implementation for operational use. Project 2: Basin-to-Global Ocean Reanalyses Canada-Mercator Contribution to CONCEPTS Validate and improve the ocean component of the basin-global modelling and assimilation system being developed for use by CONCEPTS and Mercator. Project 3: Regional Ocean Prediction:C-NOOFSCanada-Newfoundland Operational Ocean Forecasting System Project 4: Sea Ice Modelling and Data Assimilation

  4. CONCEPTS: PROJECTS OF INTERESTS • 1. Ocean Data assimilation, GOAPP (Keith Thompson) • Spaceborne Ocean Intelligence Network (Darryl Williams) • DRDC (A) Requirements and Initiatives (John Osler) • 4. Coupled models for the Gulf of St. Lawrence • (Lefaivre & Pellerin)

  5. NEMO Applications in Canada • Global -- BIO, RPNE, Dalhousie, U Quebec Montreal • North Atlantic -- BIO, Dalhousie • North Pacific -- IOS, Royal Military College • Arctic and CAA -- BIO-CIS-Mercator, U Alberta • North-western Atlantic -- CNOOFS, Dalhousie, Memorial • Gulf of St. Lawrence-Scotian Shelf-Gulf of Maine -- BIO, Dalhousie • Gulf of St. Lawrence -- BIO, U Quebec Rimouski • Great Lakes -- NWRI/RPNE, BIO

  6. CONCEPTS Project 1 & 2 Validation and analysis of the ¼-deg global NEMO-CONCEPTS* ocean model Francois Roy(1), Youyu Lu(2), Jean-Marc Belanger(3), Hal Ritchie(3), Greg Smith(4) 1- Canadian Meteorological Centre, Environment Canada 2- Ocean Sciences Division, Bedford Institute of Oceanography, Fisheries and Oceans Canada 3- Meteorological Research Division, Environment Canada 4- Biological and Physical Oceanography Section, Northwest Atlantic Fisheries Centre, Fisheries and Oceans Canada A partnership of EC, DFO, DND, GOAPP and MERCATOR-OCEAN

  7. Two global ocean configurations for core project 1: ¼-deg; 1-deg Status: ¼-deg tested for 6-yr spin-up; 10-day forecast; 1-yr simulation 1-deg model tested for multi-decade simulations; 1-deg model being coupled to 100km GEM; Validation and process studies

  8. CMC global ¼ deg 1 year model run • NEMO3-LIM2 - ¼ deg ORCA025 - 50 vertical levels (1 m surface layer) • Atmospheric forcing from global GEM at 35 km resolution • (Ua,Va,Ta,Ha,SW,LW) • CORE bulk formulas from Large and Yeager (2004) • Initial conditions: MERCATOR-OCEAN PSY 3V2 • (T,S,U,V, April 18, 2007) • SSS restoring to yearly climatology + No SST restoring • 1 Year simulation 1 (S3): 3-hourly forcing • 1 Year simulation 2 (S24): 24h-av. forcing

  9. 1 Year simulation: sensitivity to atmospheric forcing SSH yearly standard deviation (using 20 outputs per day, close to hourly) S3 S3 – S24 3 hour wind forcing - 24 hour wind forcing

  10. Yearly simulations: sensitivity to atmospheric forcing Yearly averaged surface circulation in ms-1 (first 5 m) S3 S3 – S24

  11. Wind energy input to oceanic near inertial motions: S3 Initial estimates based for April 18-30, 2007 High energy flux associated with synoptic storms • Further analyses to reveal seasonal cycle • Comparison to be made with semi-analytic solution

  12. Theme 1 and 2: Conclusion NEMO3-LIM2 with GEM forcing produces a realistic seasonal cycle without major SST drift A non-persistent warm bias occurs during the summers of the north and south hemispheres (slightly reduced with 3-hourly forcing) Significant variability in SST is added with 3-hourly forcing (diurnal cycle) SSH variability is increased in coastal areas with 3-hourly forcing and the distribution of meso-scale eddies is changed

  13. Future work Verification with in situ and satellite derived data Work to explain the summer warm bias More global diagnostics Sensitivity tests to flux parameterization (CORE, GEM physics, …) and TKE parameters Adding MERCATOR data assimilation system Pseudo-operational 10 day forecasts using GEM forcing Two-way coupling of NEMO with GEM referenced to PSY3V2

  14. 1º Global Ocean Model & Decadal Simulation Model: Horizontal: Global tri-polar grids; Nominal resolution 1° in lat/long; Meridional refinement in tropics; Vertical: 46 vertical NEMO 2.3 GOAPP Simulations: 10-yr spinup with CORE Normal Year forcing CONTROL: CORE forcing 1958-2004 HEAT: wind stress set to Normal Year WIND: buoyancy forcing set to Normal Year MJO: wind stress = Normal Year + MJO

  15. Sea-Level Trend 1993-2004 (m/yr) Obs Model ¼ o

  16. Forcing sensitivity Wind 1/4º model Heat

  17. Inter-Annual Sea-Level RMS 1993-2004 (m) 1/4º model Obs Model

  18. Forcing sensitivity Wind 1/4º model Heat

  19. Summary • Global 1º model able to reproduce large-scale SSH changes during altimeter era • Wind stress changes are primary cause of SSH trend and RMS at low and mid latitudes; • impacts of buoyancy forcing mainly show at high latitudes • 1º NEMO is being coupled to 100 km GEM; expected to be a good tool for study/prediction of intra-seasonal/seasonal variations

  20. C-NOOFS

  21. C-NOOFS system V0 MERCATOR Global Output Weekly CMC WIND NEMO v2.3 Initial State C-NOOFS T,S,U,V,W Fixed BC’s Every 24 hr Data products / plotting Webpage www.c-noofs.gc.ca

  22. C-NOOFS system V1 MERCATOR Global Output CMC WIND CMC T,Q, QSW,QLW, precip Weekly 1/4 1/12 th NEMO LIM2* BULK NEMO LIM2* BULK RESTART FILE Monitoring Diagnostics Time varying BC’s T,S,U,V,W, ice, SSH T,S,U,V,W, ice, SSH Validation M vs M M vs O Data products / plotting Ocean View Of The day Obser vations Monitoring Validation www.c-noofs.gc.ca

  23. 1/4 1/12 Observations Bonavista Bay line 2004 07 CTD stations~20 km Effects of resolution on Model and Observation Sea Island line 2008 12 19 CTD ADCP (hr)

  24. C-NOOFS Overview

  25. Observations to be assimilatedV2 • Sea level anomalies : • AVISO SSALTO/DUACS • Jason (2cm), ENVISAT and GFO (3.5cm) • Near-coast representivity error • Mean dynamic topography : • Rio et al., 2005 (~5cm rms error) • Sea surface temperature : • NCEP SST RTG (~0.7C) • In situ profile T and S data : • E.g. Argo, XBT, TAO, CTD, … • CORIOLIS (Brest) • Quality Controlled by CLS (Toulouse)

  26. Forecast Viewer http://www.c-noofs.gc.ca/viewer/ • Open access to archive of static images of SSH and 3D T,S and currents

  27. CNOOFS Dynamic Quick-view Webpage www.c-noofs.gc.ca:8080/ncWMS/godiva2.html • Uses Godiva2 viewer (from RESC) • Uses OpenLayers to quickly and easily visualize data • User-defined options for colour scale and animations • Capability to create overlays in kmz format for use in geobrowsers (e.g. GoogleEarth)

  28. C-NOOFS Monitoring/Validation System • Evaluation against: • AVISO satellite altimetry (SSH) • CMC Sea surface temperature analyses • CMC sea ice concentration analyses • Argo and other in situ data (coming soon…) • For each data type: • maps of differences • RMS error versus forecast lead time • Mean error versus forecast lead time

  29. Evaluation NWA 025 V1 against AVISO SSH for May 13, 2009 RMS difference 1-day lead time 3-day lead time 6-day lead time 10-day lead time Mean difference

  30. Evaluation NWA 025 V1 against CMC SST for May 13, 2009 1-day lead time 3-day lead time RMS difference 6-day lead time Mean difference 10-day lead time V0 V1 PSY3 PSY3

  31. Future developments and plansC-NOOFS 1-2 • Implementation of NWA12-v1 to be done this summer • Optimization of SAM2v1 assimilation system for Northwest Atlantic: • Ability to deal with tides • High-resolution SST track data • Improvement of error modes for Labrador Sea • Assimilation of seal and other additional in situ data sources • Sea ice assimilation • Produce analyses daily

  32. Future developments and plansC-NOOFS 2-2 • Model improvements: • Update to NEMOv3.2 • Add tides (variable volume) • Upgrade ice model (LIM2-EVP, LIM3 or CICE) • Optimization of ocean and ice physics • Detailed validation studies, e.g.: • Iceberg tracking • Drifters • Argo, seals, … • All in situ data we can put our hands on! • Observation quality control

  33. Sea-Ice Data Assimilation Project • Project Goal: to develop an automated ice analysis system for: • Canadian Ice Service (CIS): ice concentration, thickness/type distribution, pressure, strength and edge, deformed ice at ~1-2 km resolution • NWP: ice concentration, thickness, albedo, surface emissivity at ~5 km resolution • Benefit from experience with variational and ensemble-based assimilation for NWP: • use variational approach: incremental 3D-FGAT (first guess at appropriate time) • Developed initial prototype analysis system using CIS ice-ocean model for Canadian east-coast  plan to port system to other models/regions: • Canadian Arctic archipelago region (IPY project, Polar-GEM) • Gulf of St. Lawrence (coupled ice-ocean-atmosphere model developed at RPN/IML)

  34. Experiments with east coast coupled ice-ocean model • Goal: evaluate 3D-Var approach relative to nudging approach as currently used at CIS • Configuration: • Assimilate partial concentration of 24 ice thickness categories from CIS daily ice charts • Tests with additionally assimilating CIS RadarSAT image analyses • Experimental period: 5 Dec 2006 - 30 Jun 2007 • Forecast verification in terms of total concentration and effective ice thickness (forecast minus observation) • Only where the 1970-2000 weekly sea ice climatology indicates a non-zero probability of ice

  35. Canadian east coast model (CIOM) • Multicategory sea-ice model coupled to the Princeton ocean model (Yao et al., JGR 2000). • Viscous plastic sea-ice rheology (Hibler, JGR 1979). • GEM model atmospheric forcing every 3 hours. • Grid resolution: 1/5º longitude x 1/6º latitude, 16 sigma levels. • Originially developed by Charles Tang at BIO 48-h forecast

  36. East coast 3D-Var experiments: Example Background Estimate (24h forecast) Observations (CIS Daily Ice Chart) Analysis (initialize next forecast)

  37. Total ice concentration error StdDev small differences among data assimilation experiments } large improvement vs. persistence }

  38. ARTIC MODEL March Sea-ice velocity for the last year of a 10 year run 1º model with OMIP forcing The model consistently overestimates ice velocities which influences the sea ice distribution. The reasons for the overestimations are not yet clear.

  39. Future developments: Usage of “nesting” approach NEMO’s AGRIF allows 2-way nesting Potentially useful for CAA, e.g., 1º global + 1/4º CAA 1/2º Arctic + 1/8º CAA (2-3 km) BIO-RPN-CIS collaboration in planning Should be coordinated with CICE & I-DA developments

  40. So Where Are We Going? • Global Coupled Ocean Atmosphere Ice Forecasting System • Regional Eastern Canada Coupled Ocean Atmosphere Ice Forecasting System • 1 way nest with MERCATOR • 10 day forecasting • Ice • Service • Agrif zoom GEM / NEMO Research Direction Downscaling Ocean Only Vs Coupled Ice Assimilation vs Altimetry Assimilation

  41. GODAE Future • International group helps focus national research & development • Opportunity for synergies • Much needed collaboration • Through Working Groups

More Related