Recent Improvement of Integrated Observation Systems in Japan

1
Recent Improvement of Integrated Observation
Systems in Japan
WMO Technical Conference on Meteorological and
Environmental Instruments and Methods of
Observation TECO-2010 Helsinki, Finland, 31
August 2010
Kenji AKAEDA Observations Division, Observations
Department JMA
2
Severe Weather in JAPAN - torrential downpour
/ typhoon -
Violent Wind
Recent Disaster Distribution (1998-2004)
Landslide
Flood
3
Overview of the Observational Systems in JMA
Upper-air Observations
GPS Observation
Surface Observation
Weather Radar
Lightning Monitoring
JMA
Disaster managers TV stations private weather
associations the nation
Weather forecast Advisory Warning etc.
Doppler Lidar
4
Surface Observation Network
Observatories
156 ?,, ? AWS
1042 (? rainfall, temperature, wind,
sunshine 476) ( rainfall,
temperature, wind, sunshine, snowfall 210)
(? rainfall
356)

Each stations locate in intervals of about 17km.
Center System (JMA Headquarters)
every 10sec
every 10min
Observatory
AWS Upgrade to observe maximum instantaneous wind
speed to monitor typhoon or gusty wind
Automated Weather Station
5

Observatories (unmanned)
Observatories
AWS
QC / HK
QC / HK
QC / HK
JMA Headquarters (Tokyo)
Data statistics/archive system
Data collection/processing system
Web Server observations entered manually
(visibility, weather, etc)
data archive
data statistics
Data Collection Server
QC /QA(non-real-time) spatial consistency, time-se
ries consistency, climate range checks
Data Sharing/Storage Server
CLIMAT
QC / QA(real-time) numeric ranges,
intra-consistency checks, to add the quality flag
to the datum
data dissemination
data processing
SYNOP BUFR
data dissemination (Publication,
CD-ROM, Internet, etc)
JMA users
JMA Computer System for Meteorological Services
Public

GTS
Users
Users (Organizations, NMHSs)
6
Upper-air Observation Network
16 (Average interval 350km)
31 (Average interval 110km)
Kushiro
GPS Radiosonde (16)
Matsue
Wind Profiler (31)
Automatic Balloon Launcher (8)
7
Data Flow in Soundings of JMA
ABL sites
SHIONOMISAKI
MATSUE
NAZE
HACHIJYOUJIMA
MINAMIDAITO JIMA
WAJIMA
Ishigakijima
To JMA HQ Sounding data Status data To
Sites Control command
KUSHIRO
At CONTROLCENTER Quality control of data
Remote-controlling of ABL.
CONTROL CENTER (JMA HQ)
DATA
MBL sites
DATA
8
Data flow and QC of Wind Profiler
Radome
Radome type
Standard type
CONTROL CENTER(JMA Headquarters)
40 N

• QC
• at site
• ground clutter removal
• spectrum width check
• migrating bird removal
• precipitation effect removal
• at Center
• quadric surface check
• vertical shear check
• manual check

every 10min
30 N
140 E
130 E
9
Comparison of Rawinsonde and Wind Profiler
Measurements
Correlation of u- and v- component measurements
for the wind profiler and the rawinsonde, both
located at Hachijyojima in 2008.
u-component avg. -0.36m/s std.
2.13m/s v-component avg. -0.02m/s std.
2.06m/s
Differences between wind profiler and rawinsonde
wind component measurements (N7514).
10
GPS Precipitable Water Observation
GEONET (GPS Earth
Observation NETwork)
AMeDAS (AWS)
GEONET is comprised of about 1200 GPS observation
sites installed by the Geographical Survey
Institute, Japan. The main purpose of GEONET is
observing crustal movement with earthquakes.
11
Impact of GPS precipitable water on NWP
3-hour cumulative rainfall predicted by JMAs
5-km MSM
12
Radar Data Flow
Hazardous wind occurrence probability up to next
1hour
Tornado Nowcasting
Radar Data
Quality Assured 3-D Doppler Velocity Data
Radar Data
Control Center
Radar Data
Lightning activity up to next 1hour
Radar Data
Radar Data
Quality Assured 3-D Reflectivity Data
Lightning Nowcasting
Rainfall intensity
up to next 1 hour
Every 5minutes
Precipitaion Nowcasting
Nationwide Radar Composite Map
Straightforward extrapolation
Every 30 minutes
Radar-raingauge Composite Map
Rainfall
up to next 6 hours
Short-term Precipitation Forecasting
Rain gauge data by AWS
Data assimilation using 4D-VAR
Messocale Numerical Model
Non-hydrostatic 5 km Model operated every 3 hours
13
Sample of Lightning and Tornado Nowcast
After 1h Lightning Nowcast
Current Situation
After 1h Tornado Nowcast
14

• Summary
• Most of the observational systems are remotely
  monitored
• and controled by central system and whole data
  are gathered
• to the central system.
• QC / QA is important and JMA combines several
  method to
• keep data quality.
• AWS improved to output maximum instantaneous wind
  speed.
• All radiosonde systems are upgraded to GPS
  sonde. JMA
• operates 8 ABL, HGS and 8 normal radiosonde
  system.
• JMA has stably operated 31 WPR for 10 years. JMA
  is now
• preparing WPR guidance material as IOM.
• GPS precipitable water data are routinely used in
  NWP.
• 16 radars out of 20 radars are upgraded to
  Doppler radar.
• Doppler velocity data are utilized in NWP and
  monitoring
• tornadoes. Radar data are also utilized for
  lightning and
• tornado nowcasting.

15
Thank you !
16
Radiosonde Data Flow
GPS satellite signal
Radiosonde signal
Surface observation data incorporated in the
upper air report
JMA Headquarters (Tokyo)
Data Storage Server QC / QA (non-real-time)

Ascent data (PTU,Wind)
Calculate of observation ( including quality
control procedures such as gross error checks )
Data Statistics/Archive System QC / QA
(non-real-time)
Descent data (PTU,Location)
JMA Computer System for Meteorological Services
CLIMAT TEMP
A/N Report (for domestic use)
TEMP message
GTS
Radiosonde stations (16)
Users
17
Example of migrating-birds echo
Fukui (47616) 19 NOV 2002
Height (km)
wind
Time (JST)
Time (JST)
18
Wind Profiler Data Flow
Profiler-3
Profiler-30
Profiler-2
10 minute values of Doppler velocity and signal
intensity being sent every 10 minutes
Profiler-1
Profiler-31
WINDAS
Data quality control and remote control of
profilers being made
CONTROL CENTER(JMA Headquarters)
Horizontal and vertical components of wind and
signal intensity being sent with BUFR code every
1 hour
C O S M E T S (JMA Central Computer)
ADESS
N A P S
10 minute data being sent every 1 hour
Every 1 hour
Every 1 hour
Hourly analysis made with Meso-scale model
Forecast made with Meso-scale model
Wind profiler data
GTS
19
Nationwide Radar Composite Map
Composite Processing
Low-Altitude reflectivity of each Radar
Z-R Conversion (Z200B1.6)
Radar-AMeDAS Processing
Quality Control
Calibration Factor of each Radar
Calibration
Combining (Maximum)
Nationwide Radar Echo Composite Data