u.s. godae: global ocean prediction with the hybrid coordinate ocean model (hycom) community effort:...
TRANSCRIPT
U.S. GODAE: Global Ocean U.S. GODAE: Global Ocean Prediction with the Hybrid Prediction with the Hybrid Coordinate Ocean Model Coordinate Ocean Model
(HYCOM)(HYCOM)Community Effort:Community Effort: NRL, U. of Miami, Los Alamos, NOAA/NCEP, NOAA/AOML, NOAA/PMEL, PSI, FNMOC, NAVOCEANO, SHOM, LEGI, OPeNDAP, UNC, Rutgers, USF, Fugro-GEOS, Orbimage, Shell, ExxonMobil
Presented by Harley Hurlburt, NRL Stennis, USAPresented by Harley Hurlburt, NRL Stennis, USA
International GODAE Steering Team Meeting XInternational GODAE Steering Team Meeting XMet Office UKMet Office UK
Exeter, UKExeter, UK14-16 Nov 200514-16 Nov 2005
Present U.S. Navy Operational Capabilities Related to GODAEViewable on the web
Operational Global Ocean Product
Inputs Run by
~1/2° 2D MVOI SST Analysis 1 IR + in situ FNMOC
1/8 MODAS SST Analysis 2 IR NAVO
1/4 MODAS SSH Analysis 2 ENVISAT+GFO+JASON-1
NAVO
1/16 global NLOM nowcast/forecast system 2
ENVISAT+GFO+JASON-1SSTFNMOC winds+thermal
NAVO
http://www.ocean.nrlssc.navy.mil/global_nlomhttp://www.fnmoc.navy.mil/PUBLIChttp://www.navo.navy.mil
1 T239 or ~1/2 for atmospheric model boundary condition (on GODAE server)2 Provide subsurface temperature Real-time altimetry via NAVO Altimeter Data Fusion Center (ADFC) NLOM: NRL Layered Ocean Model GODAE: Global Ocean Data Assimilation Experiment FNMOC operates a GODAE data server with data and products from a variety of sources, including real-time altimetry from the NAVO ADFC
Global Product Mid-Lat
Resolution
Vert. Coord.
Inputs Run By Target
Date
1/8 NCOM 1 15 km /z NAVO 2005
1/32 NLOM 2 3.5 km Layered SSH, SST, NAVO 2005
1/12 HYCOM 7 km //z hydro, FNMOC NAVO 2007
1/4 HYCOM 3 20 km //z NOGAPS FNMOC 2009
1/25 HYCOM 3.5 km //z atmospheric NAVO 2011
Semi-operational Product 4 forcing
1/12 Atl. HYCOM 5 7 km //z NAVO 2006
1/12 Pac. HYCOM 7 km //z NAVO 2006
1/25 Black Sea HYCOM 3.2 km //z NAVO 2006
U.S. Navy Future Operational Transitions Related to GODAEParticipants: FNMOC, NAVO, NRL, ONR, Univ, Contractors
1 High vertical resolution for mixed layer prediction. Assimilates SSH from NLOM. Running in real-time, see http://www.ocean.nrlssc.navy.mil/global_ncom2 Running in real-time, see http://www.ocean.nrlssc.navy.mil/global_nlom3 For coupled ocean-atmosphere prediction. 4 To give NAVO/Navy experience with HYCOM without official operational status; to be replaced by global HYCOM including the 1/25 Black Sea HYCOM5 Under the National Ocean Partnership Program (NOPP), 1/12 Atlantic HYCOM demo is already running in near real-time. Includes the Mediterranean Sea. Results at http://hycom.rsmas.miami.edu/ocean_prediction.html (80Tb LAS server soon)
User Interest in Real-time Global Ocean ProductsUser Interest in Real-time Global Ocean ProductsNRL Oceanography Division Web Site Hit Statistics during 2004NRL Oceanography Division Web Site Hit Statistics during 2004
Top
25
Cou
ntr
ies
an
d #
Hit
sTop
25
Cou
ntr
ies
an
d #
Hit
s
Includes the following real-time globalIncludes the following real-time globalOcean products and other resultsOcean products and other resultsAltimeter dataAltimeter dataMODAS SSH & SST analysesMODAS SSH & SST analysesOcean prediction systemsOcean prediction systems 1/161/16 global NLOM global NLOM 1/81/8 global NCOM global NCOM 1/121/12 Atlantic HYCOM Atlantic HYCOM
Total # hits 18,467,717
Avg hits/day 50,458
# hits used in country breakdown 18,289,089
# countries with 1000 hits 67
# countries with 100 hits 120
Total number of countries 182
United States 15,917,868 Japan 971,377 China 197,225 Taiwan 137,623 Germany 97,609 Great Britain 75,999 Spain 73,888 Canada 66,436 Vietnam 65,704 New Zealand 64,092 South Korea 48,663 Greece 48,087 France 46,755 Australia 45,925 Russia 38,601 Italy 30,435 Peru 28,181 India 27,794 Mexico 25,764 Netherlands 25,081 Switzerland 24,827 Sweden 22,260 Puerto Rico 17,683 Philippines 13,558 Portugal 11,350
.08° Global HYCOM snapshot: SSH and ice (gray).08° Global HYCOM snapshot: SSH and ice (gray)
Climatological ERA15 forcing - HYCOM σ2* with 32 layers and GISS mixed layer
May 2 model year 8
.08° Global HYCOM Mean Gulf Stream .08° Global HYCOM Mean Gulf Stream And Kuroshio PathwaysAnd Kuroshio Pathways
Mean over four model yearsMean over four model years
ERA15 climatological wind & ERA15 climatological wind & thermal forcingthermal forcing
Global SSH VariabilityGlobal SSH Variability
Oct 92 – Nov 98 SSHvariability based on T/P,ERS-1 and ERS-2 altimeters(from Collecte, Localisation,Satellites (CLS))
SSH variability from .08°global HYCOM σ2* withclimatological wind and thermal forcing
Velocity Cross-section Along Luzon StraitVelocity Cross-section Along Luzon StraitSb-ADCP data (top) vs. .08° global HYCOM (bottom) in the upper 300 mSb-ADCP data (top) vs. .08° global HYCOM (bottom) in the upper 300 m
Section along 21°N between 118.5°E and 124.0°ESection along 21°N between 118.5°E and 124.0°E
Sb-ADCP data from Liang et al. (2003, DSR Pt. II)Mean from HYCOM with ERA15 wind and thermal forcing
No ocean data assimilation in HYCOM
119°E 124°E123°E122°E121°E120°E
100
200
300
100
200
300
Cross-section overlaid onmean currents and speed
m/s
NOAA/NCEP Goals for Ocean Prediction
Eddy-resolving Atlantic and Pacific Ocean prediction systems with high resolution along the U.S. coast and ~1 week forecasts
[Participation in HYCOM NOPP GODAE project led by D.B. Rao and Carlos Lozano]
- Emphasis on US coastal regions, Loop Current, Gulf Stream - Seamless boundary and initial conditions to regional ocean and bio-geo-
chemical models
Coupled atmosphere-ocean-wave for hurricane forecasts - WRF-HYCOM-WAVEWATCHIII (2007)
Eddy-permitting (1/4 or matching scale) world oceans forecast system for
medium range (up to 6 weeks) - Boundary conditions for basin-scale ocean models - SST estimate to support NWP (regional and global) - Coupled global atmosphere – ocean forecast system
NOAA/NCEP Plans for Operational Ocean Prediction Systems Related to GODAE
System
Resolution1 Vert.
Coord.
Inputs2 Target Date
Atlantic HYCOM3 4-17 km //z 2005 Pacific HYCOM 4-18 km //z
SSH, SST, hydro, NCEP/GFS forcing 2007
1 Curvilinear grid with highest resolution along the US coastline. 2 Data assimilation
- 3D Var with separable horizontal and vertical covariance for SSH, T and S - 2D Var SST analyses are nudged into the model-estimated mixed layer
3 Running daily in real-time since 20 Sept 2005 with results on the web at http://polar.ncep.noaa.gov/ofs/experimental
http://polar.ncep.noaa.gov/ofs/experimental/
Nowcast Velocity field at 50 m depth superposed on the Temperature field from May 11 – June 13
NOAA/NCEP Atlantic Ocean Forecast SystemTide Gauge Comparisons for Hurricane Katrina
Pensacola, FL
DauphinIsland, AL
Waveland, MSOcean Springs, MS
Pilot’s Station,
LA
28 Aug 0500:00
29 Aug 0500:00
30 Aug 0500:00
31 Aug 0500:00
28 Aug 0500:00
29 Aug 0500:00
30 Aug 0500:00
31 Aug 0500:00
28 Aug 0500:00
29 Aug 0500:00
30 Aug 0500:00
31 Aug 0500:00
28 Aug 0500:00
29 Aug 0500:00
30 Aug 0500:00
28 Aug 0500:00
29 Aug 0500:00
30 Aug 0500:00
Days
DaysDays
Days
Days
WFS ROMS SST and surface velocity is shown inside the dashed line and outside of this area is the NAT HYCOM model.
Warm water is detached from the Loop Current and transported northward as mesoscale eddies and filaments.
From A. Barth, A. Alvera-Azcárate, From A. Barth, A. Alvera-Azcárate, R. He, R. H. WeisbergR. He, R. H. Weisberg
University of South FloridaUniversity of South Florida
Consistency NAT Hycom and WFS Consistency NAT Hycom and WFS ROMSROMS
Velocity at station C15 (10m isobath). Velocity at station C15 (10m isobath). Due to a better representation of the coast geometry, the WFS Due to a better representation of the coast geometry, the WFS
ROMS model provides better results than the NAT HYCOM. The ROMS model provides better results than the NAT HYCOM. The OI Winds also improve the model accuracy.OI Winds also improve the model accuracy.
From A. Barth, A. Alvera-Azcárate, R. He, R. H. WeisbergFrom A. Barth, A. Alvera-Azcárate, R. He, R. H. WeisbergUniversity of South FloridaUniversity of South Florida
Comparison with Comparison with in situin situ velocity velocity
HYCOM, 29-31Aug 2005 Stennis, MS
UNC-SAB modeling system sequence that nests the regional-scale QUODDY implementation (middle) within the 1/12 deg operational HYCOM-GODAE model (left). The limited-area QUODDY implementation (right) includes the estuary and tidal inlets along the Georgia/South Carolina coast and extends to the shelf-break.
SAB Nested Finite Element Model
From Brian Blanton, University of North Carolina-Chapel Hill