Title: Hernan G' Arango
1Coupled Atmospheric-Ocean Forecast Experiments on
the New Jersey Shelf
by Hernan G. Arango Institute of Marine and
Coastal Sciences Rutgers University New
Brunswick, NJ FNMOC, December 3, 2001
2Collaborators
IMCS, Rutgers University Louis A. Bowers
(bowers_at_arctic.rutgers.edu) Rob Cermak
(cermak_at_sfos.uaf.edu) Scott M. Glenn
(glenn_at_imcs.rutgers.edu) Dale B. Haidvogel
(dale_at_imcs.rutgers.edu) Sage Lichtenwalner
(sage_at_arctic.rutgers.edu) Oscar Schofield
(oscar_at_imcs.rutgers.edu) John Wilkin
(wilkin_at_imcs.rutgers.edu) Naval Research
Laboratory, Monterey James D. Doyle
(doyle_at_nrlmry.mil) Shouping Wang (wang_at_nrlmry.mil)
3Outline
- Rutgers Long-Term Ecosystem Observatory (LEO)
- Observations
- Atmosphere-Ocean Models
- Data assimilation
- Sensitivity to Initialization and Forcing
- Ocean-Atmosphere Nowcasting/Forecasting Cycles
- Case Studies
- Sea Breeze, Cycle 2, July 2001
- Downwelling Event, Cycle 3, July 2001
- Noreaster, Cycle 6, July 2001
- Metrics
- Conclusions
4 The Observations
5LEO-15 LEO NJSOS
RUTGERS
THE STATE UNIVERSITY OF NEW JERSEY
Station
Field
3km x 3km 1996-Present
6Observational Network
7LEO-15 LEO NJSOS
THE STATE UNIVERSITY OF NEW JERSEY
Best Observation Example
RUTGERS
8 The Models
9Navy Products
NOAA, Rutgers, FERI
Global Atmospheric Forecasts
NOGAPS
NCEP
Large Scale Atmospheric Forcing
Large Scale Atmospheric Forcing
COAMPS 6 km 30 min
RAMS 4 km 30 min
Local Atmospheric Forecasts
Atmospheric Forcing
Atmospheric Forcing
Atmospheric Forcing
Wave Ocean Models
ROMS 1 km 30 min
WAM or WaveWatch3 5 km 30 min
Waves
Currents
Large Scale Ocean
Ocean Biological Models
MODAS
EcoSim
10COAMPS Coupled Ocean Atmosphere Prediction System
Operational - MEL/FNMOC
Experimental - NRL-MRY
- 27 km Spatial Resolution
- 6 hr Temporal Resolution
- 48 hr forecast issued twice daily
- Year Round
- 6 km Spatial Resolution
- 30 min Temporal Resolution
- 72 hr forecast issued twice weekly
- Summer 2001
11Experimental COAMPS Grids
- Regional Model
- Horizontal Resolution Triple-nested
54 km x 18 km x 6 km - Vertical resolution approximately 30 Sigma
levels
54 KM
Wind time-series at LEO from 6 km Experimental
COAMPS Forecast
18 KM
6 KM
12COAMPS AT LEOPressure and Wind Fields July 27,
2001 00Z
Operational
Experimental
- Data provided via FTP from MEL (Master
Environmental Laboratory) - 27 km spatial resolution
- 6 hour temporal resolution
- Regional Synoptic/Mesoscale Model
- Data provided via FTP from NRL-MRY
- 6 km spatial resolution
- 30 min temporal resolution
- Regional Mesoscale Model
13meters
ROMS Bathymetry
14meters
ROMS Bathymetry
15meters
Node A
ROMS Bathymetry
16(No Transcript)
17 Data Assimilation
18Main ESSE Components
19ESSE Flow Diagram
DE0/N
DP0/N
-
-
Most Probable Forecast
Synoptic Obs
A Posteriori Residules dr ()
Historical, Synoptic, Future in Situ/Remote
Field/Error Observations d0R0
-
-
Data Residuals
Measurement Error Covariance
d-CY(-)
Ensemble Mean
eqYj(-)
Gridded Residules
Y(-)
-
j1
Y()
Y()
Y1 Yj Yq
-
Y1 Yj Yq
0
-
E(-) P(-)
-
0
-
/-
E0 P0
0
jq
uj(o,Ip) with physical constraints
Continuous Time Model Errors Q(t)
Key
Ea() Pa()
E() P()
Field Operation Assumption
20CODAR Normalized Dominant Error Covariance
x10-2
21 Sensitivity to Initialization and Forcing
22July 1999 - Tuckerton Winds
23July 14, 1999 - 1500 GMT Temperature Cross
Sections
25 23 21 19 17 15 13 11
0 5 10 15 20 25
Temp (oC)
Depth (m)
Observations
0 5 10
15 20
Distance (km)
25 23 21 19 17 15 13 11
25 23 21 19 17 15 13 11
0 5 10 15 20 25
0 5 10 15 20 25
Depth (m)
Depth (m)
Temp (oC)
Temp (oC)
RAMS Forcing
COAMPS Forcing
0 5 10
15 20
0 5 10
15 20
Distance (km)
Distance (km)
24AVHRR SST July 28, 1999 0800 GMT
COAMPS Forcing July 28, 1999 2100 GMT
28 26 24 22 20 18 16 14 12 10
3942N
3940N
3930N
3930N
3920N
3918N
3910N
7424W 7412W 7400W
7348W
7420W 7410W 7400W
7350W
Surface Currents and Temperature (oC)
25AVHRR SST July 28, 1999 0800 GMT
28 26 24 22 20 18 16 14 12 10
MODAS/RAMS Forcing July 28, 1999 2100 GMT
3942N
3940N
3930N
3930N
3920N
3918N
3910N
7424W 7412W 7400W
7348W
7420W 7410W 7400W
7350W
Surface Currents and Temperature (oC)
26AVHRR SST July 28, 1999 0800 GMT
28 26 24 22 20 18 16 14 12 10
RAMS Forcing July 28, 1999 2100 GMT
3940N
3940N
3930N
3930N
3920N
3920N
3910N
3910N
7420W 7410W 7400W
7350W
7420W 7410W 7400W
7350W
Surface Currents and Temperature (oC)
27ROMS/COAMPS ROMS/RAMS
MODAS COAMPS
28- Ocean-Atmosphere
- Forecasting Cycles
29 11 12 13
14 15 16
17
30 Sea Breeze Cycle 2 July 17-18, 2001 0000 GMT
31The New Jersey Sea Breeze
- Develops late afternoon during weak synoptic flow
- Sea Breeze front moves inland till just after
sunset - Front observed by NWS Doppler Radar
Offshore Winds
Sea Breeze Front
Onshore Winds
32Cycle 2 Sea Breeze Mesoscale Event July 17
0000 GMT, 48 hour forecast
EXP 48 hr forecast
OP 48hr forecast
NWS Radar 07/17 22Z
33Downwelling Cycle 3 July 18-20, 2001 0000 GMT
34(No Transcript)
35(No Transcript)
36ROMS
37(No Transcript)
38Noreaster Cycle 6 July 31, 2001 0000 GMT
39Cycle 6 Noreaster Synoptic-Scale Event
- Biggest Storm of the Experiment
- 30 knot ENE Winds
- 10-12 ft Seas
- Heavy Rains
EXP 48 hr forecast
OP 48 hr forecast
OP Analysis
31 Jul 2001 0000
31 Jul 2001 0000
31 Jul 2001 0000
40The Metrics
41Atmospheric Model Validation
- Time Series Comparisons
- - Observations
- - Experimental COAMPS (EXP)
- - Operational COAMPS (OP)
- - Operational NOGAPS (NGP)
- - RUC Analyses
- Variables
- - Wind Speed and Direction
- - Barometric Pressure
- - Relative Humidity
- - Temperature
- Locations
- Over water -NOAA Buoys
- Over Land - NWS ASOS
- At the coast - Tuckerton
- Products
- - Plots
- - RMS Difference
- - Average Error
http//marine.rutgers.edu/cool/hycode2/forecast.ht
ml
- CYCLE 3 July 18 - July 21, 2001
- Wind Vectors Tuckerton 44009 BLM ACY
WWD - Temperature Tuckerton BLM ACY PHL EWR
42FAIR
Seabreeze
Noreaster
FAIR
Heat Wave
Seabreeze
Noreaster
Heat Wave
43FAIR
Seabreeze
Seabreeze
FAIR
Noreaster
Noreaster
Heat Wave
Heat Wave
44Heat Wave
Heat Wave
FAIR
FAIR
Seabreeze
Seabreeze
Noreaster
Noreaster
45FAIR
FAIR
Heat Wave
Heat Wave
Seabreeze
Seabreeze
Noreaster
Noreaster
46Tuckerton, NJAll Forecast Cycles (1-8) RMS Errors
47Ocean Model Metrics
- Thermistor strings
- Two-layer system
- Temperature
- Surface
- Top average
- Bottom average
- Thermocline
- Slope
RMS0.6865
RMS0.7972
RMS1.4863
RMS3.7447
RMS1.2795
48Cycle 3, Temperature Time Series COOL2
Thermistor
49Cycle 3, Temperature Time Series COOL2
Thermistor
MY2.5
50HR COAMPS / ROMS
LR COAMPS / ROMS
KPP
MY2.5
51Conclusions
- High spatial and temporal resolution atmospheric
forcing - produced better ocean forecasts than coarser
atmospheric - forcing.
- There is a need for a fully coupled,
high-resolution ocean- - atmosphere system.
- In shallow coastal regions, the success of ocean
data assimilation - is affected by the accuracy of atmospheric
forecasts, specially - during high wind events.
- Is very difficult to evaluate a forecast with a
single number.
52Future Work
- Multiple levels of nesting.
- Better vertical mixing parameterization?
K-epsilon? - Ensemble forecasting via optimal perturbations
using tangent - linear and adjoint techniques.
- 4D variational and IOMS data assimilation.
53R.U. C.O.O.L.
54R.U. C.O.O.L.
55For more information contact Hernan G.
Arango Rutgers University, New Brunswick,
NJ 732-932-6555 x266 arango_at_imcs.rutgers.edu h
ttp//marine.rutgers.edu/cool http//marine.rutger
s.edu/po/models/roms/index.html