GOOS Regional Forum

1
National Forecasting System (MOVE/MRI.COM) and
Japan Working Team (JWoT.GOV) Masa Kamachi
Japan Met. Agency/ Met. Res. Inst. Toshiyuki
Awaji Kyoto Univ., JAMSTEC/GODAC
1. Japan Working Team of GOV (JWoT.GOV) 2.
MOVE/MRI.COM 3. JAMSTEC (related) systems
2
Contents

• Japan Working Team of GOV (JWoT.GOV)
• Outreach (Japan) Ocean Data Assimilation Summer
  School
• from 1997-

3
Japan Working Team of GOV (JWoT.GOV)
ver. 2009/06/07
Group Systems name method Kyoto Univ. Jpn. Mar. Sci. Foundation KU-JMSF 4DVAR Jamstec (DRCMES Data Res. Center for Mar. Earth Sci.) Kyoto Univ. K-7 system Coupled 4DVAR Jamstec (Appl. Lab.) Tokyo Univ. J-COPE2 3DVAR Kyushu Univ. (RIAM) RIAMOM Kalman Filter Japan Fisheries Agency Fisheries Research Institute JADE ROMS-FRA Kalman Filter, Oi, 3DVAR JMA MRI MOVE/MRI.COM-WNP (JMA - operation MRI - research.) 3DVAR/4DVAR JMA MRI MOVE/MRI.COM-G (JMA- operation MRI - research) 3DVAR
Aim Ocean Weather Coastal prediction Climate Pacific-reanalysis (1993-2004) Model improvement. 90s ElNino Ocean Weather Variability Ppredictability of Kuroshio Ocean Weather Japan Sea Predictability Oil spill Jelly fish Ocean Weather in the Western North Pacific Predictability Jelly fish Ocean Weather Kuroshio, Oyashio, Western North Pacific Variability Predictability Nowcasting-Forecasting of ocean state, oil spill, sea ice Reanalysis (1985-2007) Climate El Nino variability Initial Condition for CGCM for El Nino seasonal forecating Reanalysis (1949-2007)
On-going Development Coastal process Environmental Pollution detection Biogeochemical process ecosystem Coupled assimilation and Decadal prediction Global warming Detection GODAC Coastal process Wind-wave interaction Regional Air-sea interaction Coastal phenomena Biogeochemical process Ecosystem Coastal prediction of physical and ecological fields OSE/sensitivity/SV Sea-ice prediction Wind-wave interaction 4DVAR Coastal disaster prevention (MOVE-Cst) (Biogeochemical process) OSE/OSSE/sensitivity Seasonal forecast Coupled assimilation (MOVE-C)
4
Contents
1. Japan Working Team of GOV (JWoT.GOV) 2. JMA
MRI MOVE/MRI.COM system (Ocean-Wea., S-I)
Recent developments Prediction of
Kuroshio Reanalysis,
prediction of oil spill (near operation)
sea ice assimilation OSE (see Task
Teams report) future system for
coastal disaster prevention
5
MOVE System (MOVE/MRI.COM)
Multivariate Ocean Variational Estimation (MOVE)
System ? Ocean Data Assimilation System in
MRI/JMA.
Model MRI.COM (Ishikawa et al. 2005)
Analysis Scheme
3DVAR with Coupled T-S EOF (Fujii and Kamachi
2003)
Model Insertion
Incremental Analysis Updates (IAU)
Observation
TS profiles Gridded SST Altimetry
6
MRI Community Ocean Model(MRI.COM)
MRI.COM Ocean General Circulation model
developed in MRI
Feature
Coordinate, Grid Arakawa B-Grid
s-Z hybrid (Free Surface
Configuration) Topographical diagonally upward
advection scheme
(Ishizaki and Motoi,
1999) Generalized Arakawa scheme (Enstrophy
conservation) Sea ice model thermodynamics

Elastic-Viscous-Plastic (EVP) dynamics
? Used in MOVE-NP and MOVE-WNP Parameteriza
tions MOVE-G Noh-Kim, Isopycnal diffusion
with GM, etc. MOVE-NP, MOVE-WNP Mellor-Yamada
L2.5, Smagorinsky, etc.
7
3DVAR Analysis Scheme in MOVE
Analysis Increment is represented by the linear
combination of the EOF modes.
Amplitudes of EOFs
Background Constraint
Constraint for T, S observation
Constraint for avoiding density inversion
Constraint for SSH observation
Seek the amplitudes of EOF modes y minimizing
the cost function J. ?Analysis increment of T and
S will be correlated. See. Fujii and Kamachi,
JGR, 2003
8
Design of background statistics
1. Vertical Correlation
Coupled T-S EOF modes (Calculated from
observed T-S profiles)
2. Horizontal Correlation
Gaussian Function (Correlation length scales
are empirically designed) No correlation
between different EOF modes
3. Model domain is partitioned into subdomains
Different Sets of EOF modes and correlation
scales are prepared for different subdomains.
9
Kuroshio Prediction with MOVE-WNP
Prediction of Kuroshio meandering in 2004
Real state (assimilation)
Growth of SSH RMSE in prediction south of Japan
Prediction
10
Comparison of the probability density function
Validation
Improvement
New Constraint
11
MOVE/MRI.COM-WNP reanalysis
Water Mass Comparison with Obs.
? Observation ? MOVE/MRI.COM-WNP Mean value in
1993 to 2005 Mean along each line (same obs.
point, depth, period) Bias in depth, density (T
S) Model bias zgt800m in Japan Sea (PM)
Reanalyses Global 1949-2008 N. Pac
1949-2008 W.N.Pac 1992-2008
Matsumoto et al., 2008
12
Prediction Experiment of Oil Spill(1997/01/02)
TOPEX/POSEIDON
MOVE/MRI.COM
COMPASS-K
13
case study 1 (2004 1/25)?
Ice in Abashiri
CNTL no ice
14
3. Routine Monitoring of Observation Diagnostics
in MOVE-G
15
Map of OBS-FG with Analysis Increment (regular
monitoring)
For the check of consistency between analysis
increment (AmF contour) and observation (OmF
color of dot) (and of distribution of the adopted
data dot).
16
Map for the equatorial Section
Example of a system problem
?
17
Map for altimetry Observation
Obs. - FG
Analysis Increment
Most of mesoscale eddies are smoothened.
18
Impact of Argo floats in MOVE-WNP (ongoing OSE)
Impact of analysis fields (100m T and S)
????
ºC
PSU
(?)
100m
Low T in subarctic
High S in subarctic
High S south of KE
High T south of KE
These are probably related to the model
bias. Impacts on the forecast will be examined.
19
Impact on the Kuroshio prediction
MOVE/MRI.COM-WNP
SSH
Prediction (Initial 01May2004)
ALL
m
NoArgo
Assimilation (Real state)
m
m
20
Comparison with ECMWF
ECMWF
AME Absolute Mean Error
Balmaseda, M. A., and D. Anderson (2008b) Impact
on initialization strategies and observations on
seasonal forecast skill. Geophys. Res. Lett.,
submitted.
21
OSE Impact on 0-6 month forecast
RMSE of NTT or NAF RMSE of ALL
RMSE Improvement ratio
RMSE of ALL
22
Summary
TAO/TRITON Array Remarkable positive impact on
NINO3 and NINO4 areas for 0-6 month SST
forecast. The Impact is not clear on western
tropical Pacific and for 7-12 month forecast.
Argo Floats Positive impact on NINO3, NINO4,
western tropical Pacific, and Western Indian
Ocean for 0-6 month forecast. The positive
impact remains for 7-12 month forecasts.
Satellite Altimetry Negative impact on the
central equatorial Pacific for 0-6 month
forecast. This negative impact may be caused
by the ignorant of the increase of the fresh
water mass.
Now developing a new scheme.
23
SV influencing on the large meander path
See Fujii et al. 2008
Initial Singular Vector (V850, T1200)
T, P at 1800m in Day25 (SV)
Cold Advection
An anticyclonic eddy is developed in the deep
layer. It will induce baroclinic instability
together with the small meander in the upper
layer.
Difference of the large meander
24
Mechanism of the meandering growth
Original state of the meander is also important
(Usui et al. 2008)
Kuroshio
(Isotherm)
Upper
Cold Advection
Baroclinic Instability
Affecting pressure
H
Down welling
It is important to reproduce the this eddy and
the Kuroshio path precisely. Altimetry data is
probably effective for that. ? We will check by
OSE.
Cold anomaly
Lower
H
H
25
Comments
Use of QC information
Large discrepancy between data and analysis is
usually originated from the system problem.
I suppose it is still difficult to use the
information for judge the data quality.
Evaluation of the observing system by OSE
For seasonal forecasts, it seem to be an only
possible way. To perform OSEs in many system is
important and still valuable for demonstrating
the importance of the observation.
Analysis and forecast sensitivity with adjoint
models
not suitable for JMA operation now (no
computational resources) Observation targeting,
calculation of the influence matrix using
singular vector decomposition may be possible on
research bases.
26
Purposes of Ocean Information Servicein JMA

• Fisheries
• Temp. front and Current at 0 - 500 m depth
• Safe and economical shipping
• SST, Strong Current and Sea Ice
• Recreation
• Temperature and current (near coast)
• Rescue activity in case of ship disaster
• SST and Current (high resolution) at surface
• Prediction of oil drift
• SST and Current (high resolution)
• Coastal flood
• Current variation and warm eddy approaching

27
Near-Real-Time information
Near-real-time information of sea surface
temperature (left) and surface current (right)
are updated every day on JMAs website.
28
30-day Forecast
Predicted SST and surface current maps (left) and
area-averaged SST time series (right) are updated
every 10-day on JMAs website.
29
High lights of Ocean Condition
High lights of ocean conditions are updated every
10-day and every month on JMAs
website. Anomalous conditions and its causes are
explained.
30
Subsurface Temperature SSH
Sea surface height is related to subsurface
seawater temperature and surface current
Monthly mean subsurface temperatures at depths
100, 200 and 400 m
31
Overview of Ocean Information Service
Oceanographic Data (SST and Subsurface
Temperature, Currents and SSH)
Forecast
Float
Prediction
Satellite
Statistical and Dynamical Prediction
R/V
BUOY
Analysis
from Met. Services, Marine Institutes,
Universities via GTS, Internet and TELEFAX.
Statistical and Dynamical Analysis
Analysis and Prediction using Computer System
Quality Check
Observation data base
32
http//goos.kishou.go.jp/
33
Future/on-going issues Sea-ice
prediction, Reanalyses (global and
regional) OSE/OSSE/sensitivity analyses Ocean
prediction with 4DVAR Coastal prediction
(coupling wind wave and OGCM) (for coastal
disaster prevention) Coupled assimilation
(NHMwaveOGCM)
34
Contents
1. Japan Working Group of GOV (JWG.GOV) 2.
JMA/MRI MOVE/MRI.COM system, Recent
developments Reanalysis, prediction
of oil spill (near operation) sea ice
assimilation future system for
coastal disaster prevention 3. JAMSTEC
Ocean Reanalysis, Coupled Reanalysis, New
Data Center (GODAC), Japan Marine Science
Foundation
35
Activity of JAMSTEC group (Ocean State Estimate
for Climate Studyby using a 4D-VAR data
assimilation system)

• 25yr ocean reanalysis experiment
• (supported by K7, DIAS project)
• Optimization Surface to 2000m-depth
• Phenomena Seasonal-interannual variability
  (ENSO, mode water)
• Research Argo impact study, OSSE, application to
  fishery etc.
• 50yr full-depth ocean reanalysis experiment
• (As 07- 08 JAMSTEC Award Research)
• Optimization Surface to bottom
• Phenomena Interannual-interdecadal variability
• (bottom water warming, PDO, ENSO
  modulation)
• Research Deep ocean sensitivity experiment,
  global heat/mass
• transport estimate, global
  warming study etc.

36
Reanalysis Dataset
Nino3 SST DMI ITF mass transport
(10-18Sv) ACC mass transport (150-160Sv) Atlant
ic MOC (16-20Sv)
Comparing with TOGA-TAO ADCP
Also, obtained 4-D velocity field is by and large
consistent with independent observations by TAO
array.
Our result provides realistic time series of
important climate indices.
37
Global Ocean 4D-VAR DA System
Optimal synthesis/Dynamical interpolation
Time change of the each component of the cost
function, i.e. the difference between simulation
and observation. Reduction by iteration processes
means progress in synthesis.
38
Reconstructed coupled fields
Ocean subsurface (thermocline tilt, warm pool,
volume transport), surface (SST, heat/freshwater
flux, wind stress), and atmospheric conditions
(vertical motions, precipitation, clouds) are
realistically reproduced as a dynamically
self-consistent field.
39
JAMSTEC Organization Chart
Research Institute for Global Change
Research Sector
Institute for Research on Earth Evolution
Institute of Biogeosciences
Major task on Earth data information
(1) Data
collection, QC, Archive, and Delivery
(2) Creation of Value
added Data by DA and so on
(3) Construction of Virtual Ocean applicable
for a variety of practical use
Leading Project
President Executive Directors
Laboratory System
Mutsu Institute for Oceanography
Kochi Institute for Core Sample Research
Research Support Department
Auditors
Development and Promotion Sector
Marine Technology Center
Earth Simulator Center
Data collection and delivery system
Data Management and Engineering Department
Global Oceanographic Data Center
Center for Deep Earth Exploration
Advanced Research and Technology Promotion
Department
Management Sector
Planning Department
Administration Department
Finance and Contracts Department
Safety and Environment Management Office
Audit and Compliance Office
40
On-demand Ocean View System by a multi-level
virtual ocean development based on 4DVAR ODA and
CDA systems on ES
For practical use of value-added datasets toward
scientific and societal benefits
41
NOWCASTING/FORECASTING SYSTEM OF THE OCEAN
CIRCULATION OFF ROKKASHO VILLAGE, Japan
Japan Marine Science Foundation

• OUR GAOLS Construction of a coastal ocean
  forecasting/nowcasting system for assessing
  radiological impact into coastal marine
  environment off the Shimokita peninsula from
  routine discharge of low-level liquid waste.
• Difficulties The target area is located between
  the subtropical and the subarctic gyre of the
  North Pacific and has a very complicated water
  mass structure. Furthermore, the Tsugaru Warm
  Current (TWC) flows into this area from the Sea
  of Japan and spreads in a variety of shapes
  seasonally.
• Methodology A system is comprised of the Kyoto
  Univ. OGCM and a one-way nesting method in order
  to satisfy needed high resolution. Products in
  the Kuroshio-Oyashio transition zone by use of
  4-dimensional variational method (4D-Var) by
  Ishikawa et al. (2008) are adopted as the lateral
  open boundary conditions. The dispersion of
  radionuclides will be estimated on the basis of
  the these flow field data by using SEA-GEARN code
  developed by JAEA.

42
Appendix
43
Information about JMAs service
a) a list of "customers" of your operational
services? Actual user (contracted user) Japan
Coast Guard (JCG), Japan Ministry of Defense
(Japan Maritime Self-Defense Force JMSDF), Japan
Meteorological Business Support Center (JMBSC)
Assumed user (1) in JMA NWP division,
Climate Prediction Division, Marine
Observatories (2) through JMA web
(general users) (3) through JMBSC
(private enterprises) (4) Others Fishery,
Shipping Agents
44
b) what type of service "products" that your
institution provides? and/or for what
applications? General current (surface and
subsurface horizontal velocity fields), SST,
subsurface temperature, salinity, sea surface
height Prediction about SAR Oil Spill (JCG,
JMA/headquarters) Surface current, surface
temperature Fishery SST, subsurface T Merchant
Service (for economical ship routing) surface
current ( wind) Marine leisure surface current
( wind) Research (physical, biological,
chemical, harbor engineering) T,S,U,V,SSH Japan
Ministry of Defense (Japan Maritime Self-Defense
Force JMSDF) subsurface T, SST, surface current
(for SAR)
45
c) (optional) what are methods of
delivery of your service products to
users/customers? (i.e., put them on the web vs.
other method of delivery) JMA web (figure and
commentary) To general users
(nowcasting and forecasting fields
SST,T,S,U,V,SSH, figure and commentary) JMA
leased line (Gridded point value GPV
Grib/Grib2 format) To Japan Coast Guard
(JCG) (nowcasting and forecasting fields,

U,V, surface wind), To Japan
Ministry of Defense (JMSDF) (nowcasting,
U,V,T,SSH), To Japan Meteorological
Business Support Center (JMBSC)
(nowcasting and forecasting fields, SST) JMA
radio facsimile broadcast (JMH) (figure)
To Japan Meteorological Business Support Center
(JMBSC) (nowcasting and
forecasting fields, SST) To ships in the
seas (nowcasting, current, T) NEAR-GOOS data
server (figure, GPV) To registered users
(nowcasting fields, of text file with the
NEAR-GOOS format) Future Plan (feasibility)
(GPV) To JMA headquarters (other
divisions) (background information
(T, U,V) to sea ice prediction model)
(SST Mixed Layer Depth to Typhoon prediction
model)