An Introduction To Himawari-8 and 9

Journal
April of the Meteorological Society of Japan, Vol.K.94,BESSHO

2016 No. 2, pp.
et 151183,
al. 2016 151
DOI:10.2151/jmsj.2016-009
An Introduction to Himawari-8/9
Japans New-Generation Geostationary Meteorological Satellites
Kotaro BESSHO, Kenji DATE, Masahiro HAYASHI
Meteorological Satellite Center, Japan Meteorological Agency, Kiyose, Japan
Akio IKEDA
Satellite Program Division, Japan Meteorological Agency, Chiyoda, Japan
Takahito IMAI
Hidekazu INOUE, Yukihiro KUMAGAI, Takuya MIYAKAWA
Hidehiko MURATA
Tomoo OHNO
Arata OKUYAMA
Ryo OYAMA
Typhoon Research Department, Meteorological Research Institute, Tsukuba, Japan
Yukio SASAKI, Yoshio SHIMAZU
Kazuki SHIMOJI, Yasuhiko SUMIDA
Corresponding author: Kotaro Bessho, Data Processing

Department, Meteorological Satellite Center, Japan Mete-
orological Agency, 3-235, Nakakiyoto, Kiyose, Tokyo
204-0012, Japan
E-mail: [email protected]
2016, Meteorological Society of Japan
152 Journal of the Meteorological Society of Japan Vol. 94, No. 2
Masuo SUZUKI, Hidetaka TANIGUCHI,
Hiroaki TSUCHIYAMA
Volcanology Division, Japan Meteorological Agency, Chiyoda, Japan
Daisaku UESAWA
Hironobu YOKOTA, and Ryo YOSHIDA
(Manuscript received 3 April 2015, in final form 7 December 2015)
Abstract
Himawari-8/9a new generation of Japanese geostationary meteorological satellitescarry state-of-the-art optical

sensors with significantly higher radiometric, spectral, and spatial resolution than those previously available in
the geostationary orbit. They have 16 observation bands, and their spatial resolution is 0.5 or 1 km for visible
and near-infrared bands and 2 km for infrared bands. These advantages, when combined with shortened revisit
times (around 10 min for Full Disk and 2.5 min for sectored regions), provide new levels of capacity for the
identification and tracking of rapidly changing weather phenomena and for the derivation of quantitative products.
For example, fundamental cloud product is retrieved from observation data of Himawari-8 operationally. Based on
the fundamental cloud product, Clear Sky Radiance and Atmospheric Motion Vector are processed for numerical
weather prediction, and volcanic ash product and Aeolian dust product are created for disaster watching and envi-
ronmental monitoring. Imageries from the satellites are distributed and disseminated to users via multiple paths,
including Internet cloud services and communication satellite services.
Keywords geostationary meteorological satellite; Himawari; satellite meteorology
(one for visible (0.500.75 m) and one for infrared

1.Introduction
(10.512.5 m)) and enabled continuous observation
The Japan Meteorological Agency (JMA) launched of atmospheric phenomena such as tropical cyclones,
its first Geostationary Meteorological Satellite thunderstorms, heavy rainfall, and heavy snow in
(GMS) in 1977 (Kodaira et al. 1978) to join a satel- the East Asia and Western Pacific regions. A series
lite constellation under the Global Atmospheric of five GMS units were launched and operated until
Research Program (GARP), an international initia- 2003. GMS-5 had an improved imager observing four
tive established by the World Meteorological Orga- bands (one for visible (0.72 m) and three for infrared
nization (WMO) and the International Council of (6.75, 11.0, and 12.0 m)), including a water vapor
Scientific Unions (ICSU) in 1967. The constellation band. The GMS series were spin-stabilized units and
consisted of the GMS and satellites from the United are generally referred to as first-generation meteoro-
States National Aeronautics and Space Adminis- logical satellites. The satellite scheduled to follow
tration (NASA), its National Oceanic and Atmo- GMS-5 was lost due to a launch vehicle failure, and
spheric Administration (NOAA), and the European the Geostationary Operational Environmental Satellite
Space Agency. Located at around 140E, the GMS (GOES)-9 was used for backup operation from 2003
was equipped with an imager observing two bands to 2005 in collaboration with NOAA/National Envi-
April 2016 K. BESSHO et al. 153
ronmental Satellite, Data, and Information Service

(NESDIS).
Two Multi-functional Transport SATellites
(MTSATs) succeeded the GMS series (JMA/MSC
2006; Miyamura 2007), fulfilling meteorological
functions for JMA and aviation control functions for
the Civil Aviation Bureau under Japans Ministry
of Land, Infrastructure, Transport and Tourism. The
MTSAT imager had five observation bands (one for
visible (0.675 m) and four for infrared (3.75, 6.75,
10.8 and 12.0 m)), and was more effective than the
GMS-5 imager in detecting low-level cloud or fog
at night. MTSAT-1R began operation in 2005, with
MTSAT-2 having a period of operation from 2010
to 2015. The MTSAT series were created as three-
axis body-stabilized satellites similar to the contem-
porary GOES series and are generally referred to as
second-generation meteorological satellites. On the
other hand, Meteosat Second Generation (MSG)
operated by EUropean organisation for the exploita-
tion of METeorological SATellites (EUMETSAT)
is a spin-stabilized unit, but is also referred to as a Fig. 1. Himawari-8/9.
second-generation meteorological satellite.
To enable uniform global earth observation,
WMO promotes the World Weather Watch (WWW) This paper provides an overview of the
program in collaboration with WMO member states. Himawari-8/9 satellites. First, the basic functions
The space-based Global Observing System (GOS), and specifications are outlined in Section 2. Section 3
which consists of geostationary meteorological satel- describes the ground segment for the new satellites. A
lites and low-earth-orbit meteorological satellites, detailed account of the new Himawari-8/9 imager is
plays an important role in the program. The GMS given in Section 4. In Section 5, JMAs Himawari-8/9
and MTSAT series have been indispensable parts of imagery and data distribution/dissemination scheme is
the GOS, providing cloud and water vapor imagery explained. Section 6 shows physical retrieval products
to more than 30 countries and territories, contributing created from satellite observation data. In Section 7,
to weather forecasts and natural-disaster counter- the development plan for new products and applica-
measures, and supporting safer transportation in the tion of Himawari-8/9 observation data are discussed.
East Asia and Western Pacific regions. The GMS and Section 8 summarizes the paper.
MTSAT series are also called the Himawari series
2. Functions and specifications
(himawari means sunflower in Japanese).
Himawari-8 was successfully launched from Himawari-8/9 have identical specifications as
Japans Tanegashima Space Center using an H-IIA listed in Table 1. They are three-axis attitude-
rocket on 7 October 2014 and settled in geostationary controlled satellites with a length of approximately 8
orbit on 16 October. JMA has commenced the units meters and a mass of approximately 3,500 kg each at
operation from 7 July 2015. Himawari-9 will also be launch. Their buses were made by Mitsubishi Elec-
launched in 2016 for in-orbit standby service and will tric, and both will be located at 140.7E. The satel-
eventually replace Himawari-8. The Himawari-8/9 lites highly improved Advanced Himawari Imagers
units are shown in Fig. 1. Both satellites will be oper- (AHIs) have capabilities comparable to those of the
ational for seven years, with Himawari-9s obser- Advanced Baseline Imager (ABI) on board GOES-R,
vation work continuing to 2029 (Fig. 2). JMAs which is run under a collaborative program between
Himawari-8/9 will enter service ahead of other NOAA and NASA (Schmit et al. 2005; Schmit et al.
third-generation meteorological satellites such as 2008). A description of the AHIs (made by Harris/
NOAA/NASAs GOES-R unit and EUMETSATs Exelis) is given in Section 4. Himawari-8/9 use the
Meteosat Third Generation (MTG) unit (Mohr 2014). Ku-band for telemetry and command operations and
Fig. 2. Himawari-8/9 schedule.
Table 1. Major specifications of Himawari-8/9.

Attitude control 3-axis stabilization
Size during operation Total length: approx. 8 m
Mass Dry mass: approx. 1300 kg
At launch: approx. 3500 kg
Design lifetime Meteorological mission: 8 years or longer
Satellite: 15 years or longer
Geostationary position 140.7 degrees east
Imager Advanced Himawari Imager (AHI)
Space environment monitor Space Environment Data Acquisition Monitor (SEDA).
Communications 1) AHI data transmission
Ka-band, 18.1 18.4 GHz (downlink)
2) Data Collection System (DCS)
Uplink from Data Collection Platforms (DCPs)
402.0 402.4 MHz (uplink)
Transmission to ground segments
Ka-band, 18.1 18.4 GHz (downlink)
3) Telemetry and command
Ku-band, 13.7514.5 GHz (uplink)
12.212.75 GHz (downlink)
Contractor Mitsubishi Electric Corporation
Launch vehicle H-IIA rocket
the Ka-band for downlinking of AHI data. Each satel- icant carrier role since the Great Sumatra-Andaman
lite also carries a transponder to relay meteorological Earthquake of 2004, the Great East Japan Earthquake
and tide/tsunami data from earth-based observing of 2011, and numerous torrential rain-related disasters
stations (Data Collection Platforms, or DCPs) in order in Japan.
to sustain the Data Collection System (DCS) of the Himawari-8/9 have a Space Environment Data
GMS and MTSAT series. JMAs geostationary meteo- Acquisition monitor (SEDA) to observe high-energy
rological satellites have taken on a much more signif- electrons and protons in orbit. JMA will internally use
Fig. 3. Overview of ground segment for Himawari-8/9.
the data obtained in this way to support satellite oper- and the Ka-band is used for data downlinks from their
ation and failure analysis. The SEDA unit, supplied by AHI and DCP units. To minimize the adverse effects
RUAG Space, is based on the EMU (Environmental of rain attenuation on these bands, antenna sites
Monitoring Unit) developed for the Galileo satellites. approximately 800 km apart are situated in the Kanto
The data it collects are provided by JMA on a near- (primary station) and Hokkaido (secondary station)
real-time basis to the National Institute of Informa- regions of Japan for diversity. Data centers are located
tion and Communications Technology (NICT), which in the vicinity of these sites to process the received
leads space weather research in Japan. data.
Each antenna site has transmitting/receiving
3. Ground segment
equipment consisting of two antennas with diam-
Himawari-8/9 and related ground equipment are eters of 9 meters, and the data centers have equip-
operated by Himawari OPeration Enterprise corpo- ment for satellite control and for processing of AHI/
ration (HOPE)a special-purpose company estab- DCP data received at the antenna sites. AHI data are
lished under a Private Finance Initiative (PFI) project. processed to level 1a data (known as Himawari radio-
HOPE controls the satellites as well as receiving data metric data, which are raw data with calibration and
derived from their AHI/DCP units and sending the navigation parameters attached), while DCP data are
data to JMA, which processes the information and formatted for JMAs telecommunications system.
provides related products to users. Figure 3 highlights The data processed at these centers are sent to JMAs
the Himawari-8/9 ground segment. Meteorological Satellite Center (MSC) and JMAs
As mentioned previously, the Ku-band is used for Osaka Regional Headquarters (referred here simply as
telemetry, tracking, and command of Himawari-8/9, Osaka). If the primary data center in Kanto should
Table 2. Himawari-8/AHI and MTSAT-2/Imager specifications.

Himawari-8/AHI MTSAT-2/Imager
Spatial Spatial
Band Wavelength Wavelength
resolution Band resolution
# (m) (m)
(km) (km)
Visible #1 0.47 1
#2 0.51
#3 0.64 0.5 VIS 0.68 1
Near-infrared #4 0.86 1
#5 1.6 2
#6 2.3
Infrared #7 3.9 IR4 3.7 4
#8 6.2 IR3 6.8 4
#9 6.9
#10 7.3
#11 8.6
#12 9.6
#13 10.4 IR1 10.8 4
#14 11.2
#15 12.4 IR2 12.0 4
#16 13.3
become dysfunctional, the secondary one in Hokkaido five bands. Table 2 shows details of the Himawari-8/
will take over its duties. AHI and MTSAT-2/Imager bands along with the
MSC and Osaka have computer systems exclu- related wavelengths, and Fig. 4 shows their spec-
sively for processing data from the centers. Himawari tral response functions (SRFs) compared with those
radiometric data received at MSC from both centers for MTSAT-2 bands. The performance of the AHI on
are interpolated, compared, and selected. Level 1b board Himawari-8 is evaluated in ground testing. The
Himawari Standard Data (HSD), High-rate Infor- test results, including data on the center wavelength,
mation Transmission (HRIT) files, and edited DCP bandwidth, bits, Signal-Noise Ratio (SNR), dynamic
data are then produced and provided to users. Osaka range, radiometric calibration accuracy, Internal
receives data only from the secondary data center Calibration Target (ICT) emissivity, and out-of-
and will issue satellite data and products if the MSC band response, can be found on the website at http://
system should become dysfunctional due to regional www.data.jma.go.jp/mscweb/en/himawari89/space_
infrastructure issues caused by natural disasters or segment/doc/AHI8_performance_test_en.pdf.
similar.
4.1 Individual band characteristics
4. AHI instruments
Each AHI band has its own characteristics. True-
The AHIs on board Himawari-8/9 are greatly color images from a combination of the three visible
improved over those of the MTSAT series in terms bands (blue: #1, 0.47 m; green: #2, 0.51 m; red: #3,
of the number of bands, spatial resolution and 0.64 m) appear as if seen by the human eye. Figure
temporal frequency. AHIs have 16 observation bands 5 shows the first true-color composite image from
(three for visible, three for near-infrared, and ten for Himawari-8 at 02:40 UTC on 18 December 2014. The
infrared), whereas the MTSAT series imagers have first images in all bands of Himawari-8 can be found
Fig. 4. SRFs for visible bands (a), near-infrared bands (b), and infrared bands (c-e) of the AHI unit on board
Himawari-8 and the Imager unit on board MTSAT-2. The solid and dashed black lines in the visible/near-infrared
band panels (a and b) represent values for AHI and Imager, respectively, and the corresponding red lines repre-
sent the same in the infrared band panels. The RGB spectra in the visible/near-infrared band panels (a and b)
were generated using the program provided at http://www.physics.sfasu.edu/astro/color/spectra.html. The black
lines in the infrared panels (c-e) represent the brightness temperatures of up-welling radiances at the top of the
atmosphere as simulated using the radiative transfer model (LBLRTM) with HITRAN2000 (AER updates) line
parameters based on the US standard atmosphere.
vapor, volcanic ash, sea surface temperature, and

other phenomena. The atmospheric window band
is present on most meteorological satellites and is
mainly used at operational weather centers to support
real-time weather analysis and forecasting. Band #16
(13.3 m), which is in the CO2 absorption band, is
used for cloud top height assignment and the estima-
tion of thin cirrus opacity.
4.2 Spatial resolutions

Table 2 also lists the AHI sub-satellite spatial reso-
lutions. The highest resolution is 0.5 km at sub-satel-
lite point for band #3 in the visible wavelength. Those
of other visible bands and band #4 in the near-in-
frared wavelength are 1 km. Other near-infrared
bands and all infrared bands have spatial resolutions
of 2 km. The values for AHI visible and infrared
bands are twice of those of MTSAT-2/Imager. Figure
6 shows the spatial resolution difference between
the visible band of MTSAT-2/Imager and band #3 of
Fig. 5. First true-color composite image from
Himawari-8/AHI at 03:00 UTC on 29 January 2015.
Himawari-8 at 02:40 UTC on 18 December
2014. From the imagery of Himawari-8/AHI, each cumulus
cloud is distinguished clearly.
on the website at http://www.jma.go.jp/jma/jma-eng/ 4.3 Observation areas and frequencies

satellite/news/himawari89/20141218_himawari8_ Figure 7 shows 10-min timelines of AHI observa-
first_images.html. Near-infrared (NIR) bands #46 tion areas and frequencies. The five scans performed
(0.86, 1.6, and 2.3 m) provide cloud physical param- by the AHIs are Full Disk, Japan Area (Regions 1 and
eter data such as information on the water/ice phase, 2), Target Area (Region 3), and two Landmark Areas
particle size, and optical thickness. Solar reflectance (Regions 4 and 5). The scan ranges for Full Disk and
differences in the visible and NIR bands enable the the Japan Area are preliminarily fixed, while those for
collection of surface property data such as informa- the Target Area and Landmark Areas are flexible to
tion on snow/ice cover and vegetation. The short- enable prompt response to meteorological conditions.
wave infrared band #7 (3.9 m), which is inherited Initial Himawari-8 operation involves the use of both
from the MTSAT series, is used to monitor low-level sets of Landmark Area data exclusively for naviga-
cloud, fog, wildfires, and other phenomena. Three tion and moon observations for calibration; the data
bands from #810 (6.2, 6.9, and 7.3 m) in the water are not intended for use in satellite products. In future
vapor absorption band are available, as opposed to the work, JMA plans to use one Landmark Area (Region
single band of the MTSAT series. The water vapor 5) for observation of phenomena such as rapidly
bands are sensitive to middle-to-upper tropospheric developing cumulonimbus cloud and to provide the
humidity, and differences in sensitivity between them resulting data to users for watching the localized
provide humidity vertical profile information. Band torrential rain brought by the cloud. On a specific
#11 (8.6 m) is used for thin ice cloud monitoring in timeline, the AHIs scan the Full Disk once, the Japan
conjunction with other atmospheric window bands. Area and Target Area four times, and each Landmark
Bands #10 and #11 are sensitive to volcanic SO2 gas. Area twenty times. Accordingly, Full Disk images
Band #12 (9.6 m), which is in the ozone absorp- are taken every 10 min, Japan Area and Target Area
tion band, is used to monitor stratospheric ozone and images every 2.5 min, and Landmark Area images
(indirectly) potential vorticity. There are three bands every 30 s. In Himawari-8/9 baseline observation,
(#1315; 10.4, 11.2, and 12.4 m) in the 1012 m the timeline is repeated every 10 min other than
atmospheric window as opposed to the two (known in housekeeping operation. Supplement 1 shows
as split window bands) of the MTSAT series. These infrared (IR1 or band #13) animations from MTSAT-2
are used to monitor ice crystals/water, lower water Northern Hemisphere images every 30 min (left)
Fig. 6. Sample of spatial resolution difference between the visible band of MTSAT-2/Imager (a) and band #3 of
Himawari-8/AHI (b) on 03:00 UTC on 29 January 2015.
Fig. 7. AHI observation areas and frequencies on a timeline of 10 min.
and Himawari-8 Japan Area images every 2.5 min at 4.4 Sample observation data and simulated data
11:0019:30 UTC on 26 January 2015. Synoptic scale To support the development of Himawari-8/9
cloud motions can be detected from the images taken products, sample observation data created from AHI
at 30-min intervals. On the other hand, mesoscale observation data and simulated proxy satellite obser-
cloud motions can be followed on the images taken vations derived from numerical modeling have been
every 2.5 min. From these animations, the advantage provided on the website at http://www.data.jma.go.jp/
of high temporal frequency of Himawari-8/AHI is mscweb/en/himawari89/space_segment/spsg_sample.
confirmed. html. The AHI observation data set is acquired in
Himawari-8 in-orbit-test period, not during its opera-
Fig. 8. Simulated images for Himawari-8s 16 AHI bands based on RSTAR.
tion. In the website, there are some source codes for mscweb/en/himawari89/space_segment/spsg_ahi_
reading and converting the sample data. proxy.html.
Figure 8 shows an example of AHI 16-band simu- For users planning to utilize only bands similar to
lated observation using a radiative transfer model those of MTSAT-2 after Himawari-8 becomes oper-
based on the provisional response functions of the ational, Himawari-8/AHI bands compatible with
imaging units. Table 3 shows radiative transfer calcu- those of MTSAT-2/Imager are selected. Bands #3, 7,
lation design based on the System for Transfer of 8, 13, and 15 of the AHIs are considered compatible
Atmospheric Radiation (RSTAR) package (Naka- with the VIS, IR4, IR3, IR1, and IR2 specifications
jima and Tanaka 1986; Nakajima and Tanaka 1988; of MTSAT-2/Imager, as shown in Table 2. In this
Stamnes et al. 1988). Analysis and forecasts from selection of compatible bands, the simulated satellite
JMAs Numerical Weather Prediction (NWP) Global observations mentioned above were used to compare
Spectral Model (GSM) (Japan Meteorological Agency Himawari-8/AHI with MTSAT-2/Imager (Murata
2013) were adopted as atmospheric fields. Since et al. 2015).
February 2013, these proxy data have been available
on the MSC website at http://www.data.jma.go.jp/
Table 3. Radiative transfer calculation design.

Radiative Transfer Model (RTM) Rstar6b (Nakajima and Tanaka 1986)
Longitude of sub-satellite point 140.0 degrees east
Spectral Response Function (SRF) Approximated with five points
Number of vertical layers in RTM 14
Atmospheric profile GSM (TL959L60, 20-km horizontal resolu-
tion) (JMA 2013)
Surface wind speed GSM
Surface reflectance MODIS product (MOD09)
Aerosol Total-column climatology value used in the
GSM
Cloud Retrieved from GSM cloud liquid/ice water
content
(Ice crystal habits: hexagonal plates, Yang
et al. 2000)
Ozone 3D climatological ozone used in the GSM
(Maki et al. 2008)
4.5 Image navigation and registration sated as attitude bias and image distortion. Band-to-
To support accurate image navigation and regis- band co-registrations are determined via direct pattern
tration (INR), the Himawari-8/9 satellites have star matching between different bands, and real-time
trackers (STTs), inertia reference units (IRUs), and correction is performed in reference to the relation-
angular rate sensors (ARSs) installed on optical ship between these co-registrations and AHI optical
benches together with their AHIs for attitude determi- system temperatures. The typical residual error after
nation. These sensors help to suppress the influence the co-registration process varies diurnally. The vari-
of thermal strain by the optical benches, and inte- ation is about 2.5 pixels for band #5 and about 0.5
grally mounted sensors on the bench reduce alignment pixels for other VIS and NIR bands. For IR bands, the
errors between them. To mitigate the thermal strain, a variation is 0.1 pixels at the largest.
carbon fiberreinforced plastic panel with low thermal
expansion is used for the optical bench. In addition, 4.6Calibration
the IRUs and ARSs are equipped close to the AHI to For detector sample calibration, the AHIs have
ensure its directional accuracy. a blackbody as the ICT and a solar diffuser as a
The INR process in the Himawari-8/9 ground Solar Calibration Target (SCT). The ICT is tempera-
processing system mainly involves the use of orbit ture-controlled using heaters and generates a constant
and attitude information. The orbit parameters are radiance. The SCT is the first solar diffuser installed
determined by ranging, and related prediction is used on any JMA geostationary meteorological satellite
for image navigation. To determine attitude, the STT imager and is made of Spectralon (manufactured by
sensors identify and track stars in the field of view. Labsphere). The MODerate resolution Imaging Spec-
The IRUs and ARSs are angular velocity sensors. The troradiometer (MODIS) of the Aqua/Terra satellites
IRUs measure drifts of attitude with variations lower has the same equipment for its on-board calibrator to
than 10 Hz, and the ARSs detect jitters higher than 10 reflect and diffuse sunlight for observation by detec-
Hz. The STTs sampling rate is 4 Hz, the IRUs is 128 tors. The required accuracy of solar calibration is
Hz, and the ARSs is 512 Hz. within 5 % radiance. According to Xiong et al. (2004),
To determine errors in the orbit and attitude estab- the expected accuracy is within 2 % reflectance and 5
lished, landmark analysis (i.e., pattern matching % radiance. The ICT is used as a hot reference along
between the database coastline and that of observa- with space view observations as a cold reference
tion images) is also used. These errors are compen- for the IR bands. Meanwhile, the SCT is used as a
bright reference for the VIS and NIR bands together to apply the Midnight Blackbody Calibration Correc-
with space view observations as dark references. tion (MBCC) algorithm (Weinreb and Han 2003) to
Using these calibration targets, linear and bias cali- offset nighttime effects of blackbody in the same way
bration coefficients are derived to convert counts of as the GOES imagers (Johnson and Weinreb 1996).
raw detector sample data into radiances. After solar However, no significant nighttime effect in the AHIs
diffuser stability checking was completed on June and their blackbodies is expected.
2015, calibration coefficients were updated based
on the observation results from SCT in the visible/ 5.Level-1B products and data distribution/
near-infrared bands. In this stability checking, JMA dissemination
plans to perform lunar observation for long-period JMA processes satellite-derived level-1a Himawari
stable references (Kieffer and Stone 2005). radiometric data to create HSD as a level-1b product
To check the quality of the data obtained, JMA has in Himawari Standard Format (HSF). HSD are used
developed visible and infrared vicarious calibration as master data for all Himawari-8/9 products. Details
methods under the framework of the Global Space- of HSD and HSF are provided on the MSC website
based Inter-Calibration System (GSICS) (Gold- at http://www.data.jma.go.jp/mscweb/en/himawari89/
berg et al. 2011) and collaborative research with space_segment/spsg_sample.html. Himawari-8/9 imagery
the University of Tokyos Atmosphere and Ocean in other data formats such as HRIT, Low Rate Infor-
Research Institute (AORI). The infrared calibration mation Transmission (LRIT), Portable Network
approach developed under the GSICS project utilizes Graphics (PNG), and Network Common Data Form
hyper-spectral infrared sounders (hyper sounders) (NetCDF) is also processed from HSD and provided
such as the Infrared Atmospheric Sounding Interfer- by JMA via multiple paths. Sample data created in
ometer (IASI) on board EUMETSATs Metop satel- various formats from AHI observation and proxy data
lite and the Atmospheric InfraRed Sounder (AIRS) can be found on the same website.
on board NASAs Aqua satellite (Gunshor 2006; Imagery from MTSAT-2 satellite is provided via
Tobin 2006; Tahara 2009). The approach for compar- MTSAT-1R direct dissemination through the L-band
ison of imagers on board geostationary satellites with frequency HRIT and LRIT services. Most National
hyper sounders involves generating a super channel Meteorological and Hydrological Services (NMHSs)
which clones the broad channel of the hyper sounder in the East Asia and Western Pacific regions receive
by minimizing the spectral response difference. This this imagery using L-band antennas and receivers and
method supports the investigation of sensor bias and process it with dedicated systems. JMA also provides
its seasonal or diurnal variation. The MSC website the same HRIT service imagery via the online JMA
(http://www.data.jma.go.jp/mscweb/data/monitoring/ Data Dissemination System (JDDS).
calibration.html) highlights calibration monitoring Himawari-8/9 do not carry equipment for direct
for Himawari-8 and the MTSAT series based on this dissemination, as detailed in Section 2. Instead, all
approach. Visible/near-infrared band calibration imageries derived from the satellites are distributed
methods are also being examined within the GSICS to NMHSs via the HimawariCloud service. JMA also
framework. JMA has further developed an approach runs the HimawariCast service, by which primary sets
for visible/near-infrared vicarious calibration as part of imagery are disseminated to users via a communi-
of collaborative research with AORI. The technique cation satellite using Digital Video Broadcasting-Sat-
is based on radiative transfer computation above ellite-Second Generation (DVB-S2) technology. This
spatially uniform targets such as cloud-free ocean section covers the distribution/dissemination paths
areas and liquid cloud top (Kosaka 2012). Informa- used by JMA.
tion on the imager sensitivity trend of Himawari-8
and MTSAT-2 is also provided on the same MSC 5.1 HimawariCloud service
website. JMA plans to validate AHI sensor charac- To distribute the enormous volumes of imagery that
teristics based on these approaches. AHI fundamental Himawari-8/9 produce, in April 2015, JMA estab-
performance in-orbit-test (e.g., radiometric calibration lished an Internet cloud service (HimawariCloud) for
accuracy and SNR) is summarized in Okuyama et al. NMHSs in the East Asia and Western Pacific regions.
(2015). Table 4 shows the data set distributed via this service.
The AHIs have temperature sensors on their mirrors HSD are provided as core data with 16 bands and
and other inner structures. The use of data from these the finest spatial resolution. The true-color images
sensors in ground segment processing enables JMA detailed in Section 4 are provided in PNG format.
Table 4. Data set distributed via the HimawariCloud service.

Format Observation area Notes
Himawari Standard Data Full Disk - Interval: 10 min (Full Disk); 2.5 min (Target
Target Area Area)
- Number of bands: 16
- Spatial resolution: VIS: 0.51 km; NIR: 12
km: IR: 2 km
Portable Network Graphics Full Disk - True-color images (composites of three
(PNG) Target Area visible bands)
- Interval: 10 minutes (Full Disk); 2.5 min
(Target Area)
- Spatial resolution: 1 km
Network Common Data Form Target Area - Interval: 2.5 min
(NetCDF) - Number of bands: 16
- Spatial resolution: VIS: 0.51 km, NIR: 12
km; IR: 2 km
Table 5. Data set continuously distributed via JDDS.

Format Observation area Notes
HRIT files Full Disk - Format compatible with the MTSAT series
HRIT service
- Interval: 30 minutes
- Number of bands: 5 (VIS: 1; IR: 4)
- Spatial resolution: VIS: 1 km; IR: 4 km
For rapid scanning observation, imagery in NetCDF 5.2 HimawariCast service

is also created and distributed. NMHSs are able to The HimawariCloud service requires high-speed
access the cloud and retrieve data using an HTTP 1.1 Internet access, whereas the HimawariCast service is
client such as a Web browser or a Wget command. suitable for users with limited Internet connections.
It should be noted that an Internet connection with a Table 6 shows the data set disseminated via commu-
speed of at least 20 Mbps is required to download all nication satellite. HRIT files represent the core data
HSD. These data are separately created for each band of the service, and LRIT files are also included to
and divided into 10 segments from north to south so support LRIT service users. The dissemination further
that NMHSs can select only the files necessary for includes meteorological data other than Himawari
their operations. imagery in SATellite Animation and Interactive Diag-
Several Japanese universities and research orga- nosis (SATAID) format. JMAs SATAID visualization
nizations have received Himawari-8/9 data via the software enables the display and analysis of satel-
HimawariCloud service and re-distributed them for lite images with other meteorological data. A sample
research purposes. These are not JMA organizations; SATAID viewer image is shown in Fig. 9. SATAID
their services are based on voluntary work and are enables the superimposition of various data and prod-
conducted on a best-effort basis. ucts, such as NWP information and in-situ observa-
To support service continuity for JDDS users, JMA tion data, onto satellite imagery. It is widely used by
also produces HRIT files compatible with MTSAT NMHSs in the East Asia and Western Pacific regions
data (Table 5). These feature five bands near the as an operational tool for daily weather analysis and
MTSAT observation bands and are provided every 30 forecasting due to its usefulness and convenience.
min with the same frequency as the MTSAT service. The HimawariCast receiving and processing system
includes a C-band antenna system, a Low Noise
Block converter (LNB), a DVB-S2 receiver, and a
Table 6. Data set distributed via the HimawariCast service.

Data type Format Notes
Himawari imagery HRIT files - Format compatible with the MTSAT series
(Full Disk) HRIT services
- Interval: 10 min
- Number of bands: 14 (VIS: 1; NIR: 3; IR: 10)
- Spatial resolution: VIS: 1 km; NIR: 4 km; IR:
4 km
LRIT files - Format compatible with the MTSAT series
LRIT services
- Interval: 10 min
- Number of bands: 4 (VIS: 1; IR: 3)
NWP products (GPV) SATAID format - JMA Global Spectral Model (GSM) products
- Interval: 6 hr
In-situ observations SATAID format - Observational data for the East Asia and
(surface stations, ships, Western Pacific regions
radiosondes)
ASCAT ocean surface SATAID format - Observational data from EUMETSATs Metop
wind polar-orbiting satellites
desktop computer with data casting client software rapid downloading of essential imagery for meteoro-
and visualization software. More information on the logical services, the Agency also formulates various
system can be found at http://www.data.jma.go.jp/ types of cut-out imagery in Joint Photographic
mscweb/en/himawari89/himawari_cast/himawari_ Experts Group (JPEG) format on its website:
cast.html. http://www.jma.go.jp/en/gms/
JMAs HimawariCast service started in January http://www.jma.go.jp/en/gms150jp/
2015 with MTSAT-2 still in operation. MTSAT-2 http://www.data.jma.go.jp/mscweb/data/himawari/
imagery was disseminated through the service in index.html
parallel with direct dissemination via MTSAT-1R
6. Level-2 products and contribution to climate
until the start of Himawari-8 operation. From July
research activity
2015, Himawari-8 data imagery has been dissemi-
nated operationally via the HimawariCast service. The imagers on board the MTSAT second-gen-
eration geostationary meteorological satellites have
5.3 WMO Information System Portal five observation bands (one for visible and four for
JMA operates the WIS Portal website as the Tokyo infrared). Using imagery from these bands, JMA
Global Information System Centre (GISC-Tokyo) of monitors synoptic-scale meteorological phenomena
the WMO Information System (WIS). Himawari-8 such as extratropical cyclones, Baiu front activity,
imagery in SATAID format and the meteorolog- and tropical cyclones to help mitigate the effects of
ical data shown in Table 7 are provided through the meteorological disasters. The best practice of these
SATAID service, which is a data distribution initiative efforts involves employment of the Dvorak method
of GISC-Tokyo. The Internet-based SATAID service to enable estimation of tropical cyclone intensity
and HimawariCast data delivered via a communica- (Dvorak 1975, 1984; Velden et al. 2006). The Objec-
tion satellite are provided on a complementary basis tive Dvorak Method was also developed by Velden
for SATAID users. et al. (1998) and Olander and Velden (2007). The
Dvorak method is used by the Regional Specialized
5.4 JMA website Meteorological Centre (RSMC) Tokyo to analyze the
JMA provides Himawari-8 imagery in PNG format intensity of typhoons based on satellite observation.
on its website for public consumption. To facilitate In 2013, RSMC Tokyo has also started the practical
Fig. 9. SATAID viewer image. Satellite infrared imagery (black and white) is visualized with contours of surface
pressure (yellow), surface air temperature (red), and wind barbs (blue) from NWP.
Table 7. Data set distributed via GISC-Tokyos SATAID service .

Data type Format Notes
Himawari imagery SATAID format - Interval: 10 min
(cut-out from Full Disk) - Number of bands: 5 (VIS: 1; IR: 4)
NWP products (GPV) SATAID format - JMA Global Spectral Model (GSM) products
- Interval: 6 hr
In-situ observations SATAID format - Observational data for the East Asia and
(surface stations, ships, Western Pacific regions
radiosondes)
ASCAT ocean surface SATAID format - Observational data from EUMETSATs Metop
wind polar-orbiting satellites
use of the Objective Dvorak Method (Kishimoto et al. Level-2 physical products are also utilized in JMAs
2013). From the start of the operation of Himawari-8, work, as exemplified by the indispensable role played
the method has been applied to its data set. To obtain by Clear Sky Radiance (CSR) and Atmospheric
maximal use from the data of new-generation geosta- Motion Vector (AMV) data from MTSAT in the
tionary satellites such as Himawari-8/9 and GOES-R, initialization of NWP. For aviation operators, JMA/
the Objective Dvorak Method is being updated MSC provides volcanic ash information from the
(Goodman et al. 2012). satellites to the Tokyo Volcanic Ash Advisory Center
(VAAC). MTSAT environmental products such as the Himawari-8/9 infrared bands (2 km at the sub-sat-
data on dust presence, sea surface temperature estima- ellite point). The algorithm is based on methods
tion, and retrieval of sea ice motion vectors also play developed by the Nowcasting Satellite Application
a significant role in JMAs operations. Facility (Meteo-France 2012). It involves the use
As described in Section 4, Himawari-8/9 offer high of a threshold technique with brightness tempera-
observation potential to support users in the develop- ture and reflectivity for cloud mask and cloud type/
ment and improvement of various physical products. phase retrieval. It also incorporates a local radiative
Previous satellite products such as cloud property, center and a threshold tuning method developed by
CSR, and AMV information have been improved NOAA/NESDIS (Heidinger 2011). The interpolation,
using AHI data. JMA also uses these data in the intercept (Szejwach 1982), and radiance rationing
development of new retrieval algorithms for products methods (Menzel et al. 1983) are applied for cloud
relating to cloud-free areas, such as data on aerosol height assignment. The fundamental cloud product
properties with a focus on volcanic ash and dust. algorithm involves the use of radiative transfer calcu-
The sections below briefly outline basic meteorolog- lation results obtained from NWP data to determine
ical products developed by JMA from Himawari-8/9 thresholds for CSR from infrared bands. A prototype
AHI data in collaboration with geostationary meteo- product was made using MSG/Spinning Enhanced
rological satellite employment organizations such as Visible and InfraRed Imager (SEVIRI) data (Schmetz
NOAA/NESDIS and EUMETSAT. JMA and these at al. 2002). Table 8 shows evaluation results for the
organizations are members of the Coordination Group prototype cloud mask, cloud top height, cloud type,
for Meteorological Satellites (CGMS), which consists and phase compared with the Aqua/MODIS Level2
of developers and operators of meteorological satel- products (MYD35_L2 and MYD06_L2) as refer-
lites, space agencies operating research and devel- ences. Evaluation was performed over periods of
oping satellites, and WMO. two weeks in each season (28 December 201110
In addition to the geophysical products listed January 2012, 28 March 201210 April 2012, 27
above, JMA contributes to international climate June 201210 July 2012, and 27 September 201210
research frameworks such as the International Satel- October 2012), and scores were calculated using
lite Cloud Climatology Project (ISCCP), the Global temporally and spatially matched MODIS products
Precipitation Climatology Project (GPCP), and the within the four periods. Accuracy and hit rate are
Sustained, Co-Ordinated Processing of Environmental defined as (A+D)/(A+B+C+D) and A/(A+B), respec-
satellite data for Climate Monitoring (SCOPE-CM) tively, and shown in Table 8 (d). Cloud mask detec-
initiatively by providing geostationary meteorological tion accuracy was 85 %, and most of the other scores
satellite observation data and products. These activi- were also reasonable. However, the evaluation results
ties are also briefly outlined below. also suggest the need for further improvement in
cloud detection over snow/ice areas and cloud height
6.1 Fundamental cloud product assignment for semi-transparent cloud.
Cloud property information is a highly essential
meteorological product derived from satellite obser- 6.2 CSR product
vations, not only because it is used for weather anal- The CSR product provides area-averaged radi-
ysis, but also because it is applied in the retrieval of ances and brightness temperatures for cloud-free
other products (e.g., CSR and AMV). For example, pixels (Uesawa 2009). NWP centers will use the
masking the cloud area is the first step in the deter- Himawari-8/9 CSR product in their operational assim-
mination of surface parameters in satellite remote ilation systems as a primary data source. Data are
sensing. Considering the importance of cloud prop- calculated for each 16 16 infrared pixel box corre-
erty information, JMA is currently operating and sponding to approximately 32 32 km2 resolution
developing two cloud products with data from at the sub-satellite point for the global data assimila-
Himawari-8/9. One is based on a statistical approach, tion system. The status of pixels as clear or cloudy is
and the other involves a physical (1D-Var) approach. determined from the fundamental cloud product. In
This section outlines the cloud product using the addition to the ordinary clear pixels provided from
former, and the latter is described in Section 7.1. this product, cloudy pixels in which the contribution
The fundamental cloud product based on the statis- of cloud top emission to total radiance is negligible
tical approach provides data on cloud mask, cloud can also be regarded as clear for AHI bands with
top height, cloud type, and phase for each pixel of strong absorption (e.g., water vapor bands). These
Table 8. Evaluation of fundamental cloud product (prototype). (a) Accuracy of cloud mask for individual cloud types and
phases, (b) accuracy of cloud mask for individual seasons and surfaces, (c) bias and standard deviation of cloud height for
individual cloud types, and (d) definitions of accuracy and hit rate.
(a)
Accuracy/hit rate
Mask (Ac.) 0.85
Type Opaque (H.R.) 0.66
Semi-transparent (H.R.) 0.80
Phase Ice cloud (H.R.) 0.98
Liquid water cloud (H.R.) 0.94
(b)
Accuracy
Mask
Winter Spring Summer Autumn All
All regions 0.86 0.85 0.85 0.85 0.85
Individual Sea 0.86 0.85 0.85 0.86 0.85
Snow/Ice 0.86 0.81 0.72 0.86 0.81
Dessert 0.87 0.86 0.86 0.89 0.87
Vegetation 0.86 0.84 0.83 0.83 0.84
Others 0.80 0.82 0.86 0.84 0.84
(c)
Height Bias (km) Standard deviation (km) Number
Low opaque 0.1 1.34 2296940
High/mid opaque 0.31 1.33 1375505
Opaque 0.05 1.34 3672445
Semi-transparent (intercept or
2.38 3.34 877586
radiance rationing)
(d)
MODIS product detection
Yes No
Fundamental cloud product Yes Correct detection (A) False alarms (B)
detection No Misses (C) Correct negatives (D)
A, B, C and D denote the number of occurrences for each case. Accuracy and hit rate are defined as
below.
Accuracy = (A + D) / (A + B + C+ D)
Hit rate = A / (A + B)
CSR products are provided in Binary Universal Form satellite products in NWP. For example, Bormann
for data Representation (BUFR) via the Global Tele- et al. (2012) showed that AMVs from satellites have
communication System (GTS). a positive impact on forecast skill in the ECMWF data
assimilation system. JMA has developed a new algo-
6.3 AMV product rithm for Himawari-8/9 AMVs (Shimoji 2014) based
AMV is one of the most important meteorological on an optimal estimation method for full exploitation
of satellite data. Improvement from the previous oper- 6.5 Aeolian dust product
ational algorithm (Oyama 2010) results in the compu- JMA monitors trends of Aeolian dust in light
tation of wind vectors with high spatial resolution of the serious damage it causes to social activity,
based on the tracking of smaller cloud patterns. Figure human health and economies in East Asia. Using
10 shows a comparison of MTSAT observation data Himawari-8/9 observation data, the Agency produces
AMVs produced with the previous operational algo- two aerosol products for Asian dust event monitoring.
rithm and the new algorithm for Himawari-8/9. Wind One of these, derived from visible and near-infrared
vectors with the new high-quality algorithm (Quality data, provides daytime information on aerosol optical
Indicator (QI) > 80) (Holmlund et al. 1998) are seen thickness and Angstrom index values (a metric of
in the figure over wide areas where none are produced aerosol particle size) for areas over oceans and also
with the previous version. As with CSR, the AMV provides aerosol optical thickness values for areas
products are provided in BUFR via the GTS. over land (Fig. 13). The algorithm involves the use
of a look-up table calculated with the assumption of
6.4 Volcanic ash product a spheroid-particle aerosol model (Okada et al. 2001).
Volcanic ash quantitative information (e.g., ash The other product provides quantitative information
cloud height, optical depth, effective particle radius, on dust using infrared data from Himawari-8/9. One
and mass loading) from satellites is especially of the advantages of the infrared dust product is its
important in aviation usage. When Eyjafjallajkull availability during both day and night time. The algo-
erupted in 2010, Stohl et al. (2011) used volcanic ash rithm is based on that of the volcanic ash product for
information from satellites as initial values for their GOES-R developed by NOAA/NESDIS mentioned
dispersion model. Frequent AHI multispectral obser- in the previous section, as the optical properties of
vation is expected to help to improve this product. dust are similar to those of volcanic ash. In addition
JMA has introduced two volcanic ash product to these two algorithms, the development of a further
retrieval algorithms for Himawari-8/9. One was algorithm is under discussion with aerosol scientists
provided by EUMETSAT (Prata 2011), and the other in Japan for aerosol data assimilation. JMA will use
was provided by NOAA/NESDIS (Pavolonis and these new aerosol products based on Himawari-8/9
Sieglaff 2012). The EUMETSAT algorithm was modi- observation data in the future.
fied by JMA, and its cloud height assignment data are
provided to Tokyo VAAC (Tsuchiyama 2013). JMA is 6.6 Rapidly Developing Cumulus Area product
also currently developing a test bed for intercompar- Himawari-8/9 make observations over Japan at
ison of the volcanic ash retrieval algorithms. The test 2.5-min intervals. The advent of frequent obser-
bed is built within the framework of the Sustained, vations from space is expected to help to improve
Co-Ordinated Processing of Environmental satellite severe weather analysis and forecasting. The Rapidly
data for NoWCasting (SCOPE-NWC) Pilot Project Developing Cumulus Area (RDCA) product has been
for comparison of the EUMETSAT and NOAA/ developed to support the monitoring of rapidly devel-
NESDIS algorithms. A case study involving volcanic oping convective cloud based on rapid-scan (5-min
ash from Mt. Kirishima on the Japanese island of interval) observations made by MTSAT-1R around
Kyushu was conducted using MTSAT-2 observa- Japan during the daytime (00 to 09 UTC) in summer
tion data as shown in Fig. 11, where the ash cloud (June to September) (Sumida et al. 2015). The RDCA
from the volcano is shown in red. Figure 12 shows product highlights areas at high risk of active convec-
the preliminary results of intercomparison for ash tion and heavy rainfall. Its algorithm detects rapid
cloud height data obtained with the two algorithms. changes in cloud conditions (e.g., sudden genesis and
A marked difference in their spatial distributions is evolution of cumulonimbus cloud) based on a statis-
observed, especially in the southwestern area. Ash tical approach by matching up satellite observation
cloud height data from the NOAA/NESDIS algorithm with ground-based lightning remote sensing obser-
are located at 06 km in the histogram, while those vation. Figure 14 shows an example of RDCA infor-
from EUMETSAT are located at 24 km. A scatter mation retrieved from MTSAT-1R observation data at
plot of the two data sets shows the same tendency. In 08:45 UTC on 11 July 2011. The green region shows
this case, there is no data set such as ground-based a rapidly developing cumulus area, and the red region
lidar for validating which product is more accurate. shows a cumulonimbus area. These areas corre-
Accordingly, further study is needed to clarify the spond closely to the severe convective region esti-
reason for this difference. mated from ground-based remote sensing results (i.e.,
Fig. 10. MTSAT IR wind vectors (QI > 80) retrieved using the previous operational algorithm (a) and with the
newly developed algorithm for Himawari-8/9 (b) at 00 UTC on 2 March 2014. Warm colors correspond to
low-level wind and cold colors to high-level wind.
those from radars and lightning detection networks every 10 min. This product will contribute to safety
in Japan). The products algorithm will be modified and control of air traffic over the East Asia and
using Japan Area observation data observed every 2.5 Western Pacific regions.
min by Himawari-8/9. After this improvement, the
RDCA product will be provided at all times during 6.7 RGB composite image product
every season. As a further step, JMA has started plan- As Himawari-8/9 have 16 bands that can be
ning the development of global RDCA product using combined in a large number of ways, it will be diffi-
Himawari-8/9 Full Disk observation data obtained cult for forecasters and analysts to digest combined
Fig. 11. RGB composite image from MTSAT-2 at 00 UTC 27 January 2011. RGB are defined (red: IR10.8IR12.0;
green: IR3.9IR10.8; blue: IR10.8) to enhance volcanic ash presence (Pavolonis 2013, personal communication).
AHI imagery at a glance. Accordingly, there is a Precipitation Climatology Project (GPCP) is to help
need for the satellite operator to provide satellite clarify spatial and temporal patterns of global precip-
images that can be easily interrupted. Red, green, and itation. Data from rain gauge stations, geostationary/
blue (RGB) composite images support this purpose. low-earth-orbit satellites, and sounding observa-
Imagery from Himawari-8/9 AHI is provided by tion work are merged to enable the formulation of
JMA for NMHSs in Southeast Asia and the South monthly rainfall grid maps. JMA has contributed to
Pacific islands via the MSC website in the combina- the ISCCP since data collection started in July 1983
tions recommended at the RGB Composite Satellite and to the GPCP since its establishment in 1986. The
Imagery Workshop, which are based on observa- Agency processes observation data from eight geosta-
tion results for MSG/SEVIRI with 12 bands (WMO tionary satellites (including GMS, the MTSAT series,
and EUMETSAT 2007). As with the MTSAT RGB and GOES-9) and provides the outcomes to ISCCP
composite images on the MSC website, Himawari-8/9 and GPCP processing centers on a routine basis. The
RGB images will contribute to the WMO Commis- resulting ISCCP and GPCP products improve under-
sion for Basic Systems (CBS) Severe Weather Fore- standing and modeling of the climate and support
casting Demonstration Project (SWFDP) and the cloud studies, including clarification of the hydrolog-
Severe Weather Forecasting and Disaster Risk Reduc- ical cycle. JMA will continue to support both projects
tion Demonstration Project (SWFDDP) (Fig. 15). after the start of Himawari-8/9 operation.
6.8 Contribution to ISCCP and GPCP 6.9 Contribution to SCOPE-CM

The International Satellite Cloud Climatology SCOPE-CM is envisioned as a component of a
Project (ISCCP) was established to collect and framework linking satellite observation technology
analyze weather satellite radiance data that would such as GOS with inter-calibration technology such
help to clarify the global distribution, properties, and as GSICS to provide required climate data records
temporal variation of clouds. Its datasets and anal- addressing the Essential Climate Variables (ECVs)
ysis products are used to study how clouds affect (Lattanzio et al. 2013). JMA computes AMVs, CSR,
climatic characteristics. The purpose of the Global and land surface albedo data from the observations of
Fig. 12. Volcanic ash cloud height retrieval intercomparison between the NOAA/NESDIS and EUMETSAT algo-
rithms for Mt. Kirishima at 00 UTC on 27 January 2011. The graphs show spatial distribution of height (a and b),
histograms (c), and a scatter plot representation (d).
past satellites and provides the results to SCOPE-CM. climate research.

AMVs and CSR products are also provided to the
7. Future plans
European Centre for Medium-Range Weather Fore-
casts (ECMWF) Re-Analysis (ERA) Interim Project The previous section outlined basic physical
(Dee et al. 2011) and the Japanese 55-year Re-Anal- products retrieved from Himawari-8/9 observation
ysis (JRA-55) Project (Kobayashi et al. 2015) for data by JMA. Each product has its own improve-
contribution to climate research fields. Himawari-8/9 ments over the previous ones developed for MTSAT
are capable of performing Full Disk observation every series data. However, there is still plenty of scope to
10 min. JMAs development of a new AMVs retrieval improve existing or develop new physical products
method for the satellites as outlined above is expected or satellite applications using observation data from
to represent an ongoing contribution to long-term Himawari-8/9. As already detailed, Himawari-8/9
Fig. 13. Prototype aerosol optical thickness product without cloud mask derived using the algorithm for
Himawari-8/9 made using data from NASAs Terra/MODIS satellite with bands corresponding to those of AHI/
Himawari-8/9.
have huge capability in terms of temporal and spatial thickness, particle size, thermo-dynamical phase, and
high-resolution observation with 16 bands. To fully cloud top temperature from AHI observation data are
utilize the data, it is necessary to keep some key considered important in climatological cloud data-
concepts in mind such as conjunction with numerical base development. Other environmental parameters
models, advanced data assimilation, application to retrieved from Himawari-8/9 AHIs are also viewed
frontier themes, and combination of multi-observation as indispensable for their coverage of wide areas and
platforms. As no operator has yet handled data sets long time periods in the fields of hydrology, glaci-
from new-generation geostationary meteorological ology, oceanography, and ecology.
satellites, innovative ideas based on these concepts In addition to their advantage of having more
are needed to bridge the gap between observation data spectral bands, the AHIs feature faster imaging and
supply and usage. higher spatial resolution than MTSAT/Imager. The
Until now, data from low-earth-orbit environ- detailed cloud information inferred from Full Disk
mental satellites and meteorological polar-orbiting images taken every 10 min and Japan-area images
satellites with visible/infrared imagers have helped taken every 2.5 min by the AHIs shows the lifecycle
to reveal major characteristics of cloud, especially in of cloud, including its genesis, evolution, and decay,
climatology, through the use of a variety of physical more clearly than ever before. By way of example,
product retrieval algorithms (e.g., Remer et al. 2005). Nakajima et al. (2010a, b) suggested that temporal
Application of this knowledge to observation results variations in cloud evolution could be clarified using
from the 16 AHI bands is expected to offer new observation results from AHI bands #57, which
insights into cloud climatology. By way of example, belong to the near-infrared or infrared wavelength.
information showing cloud properties such as optical Cloud droplet size can also be estimated from differ-
Fig. 14. RDCA product imagery at 08:45 UTC on 11 July 2011. RDCAs are shown in green, cumulonimbus in red,
and unknown areas covered by mid- or low-level cloud in aqua. The nighttime area where the solar zenith angle
exceeds 75 is shown in purple; no RDCAs are seen in the nighttime area.
ences in the effective vertical weighting function Exploration Agency (JAXA) and is tasked with
between these bands. Traditional radar meteorology, discussing the needs of satellite data users, the direc-
which involves scientific study of the atmosphere tion of research and development in satellite usage,
using radar observation results, can be seen as a and the promotion of satellite data utilization. In 2014,
type of mesoscale rain meteorology. The AHI units two working groups were organized under the panel
on board Himawari-8/9 will open the door to a new to exploit techniques for Himawari-8/9 data applica-
era of mesoscale cloud meteorology. In other words, tion in science relating to the atmosphere and oceans/
mesoscale satellite meteorology will be born from land. JMA pursues the development of new physical
these new-generation geostationary meteorological products from Himawari-8/9 in collaboration with the
satellites. The next step in this satellite revolution will scientific community in Japan as outlined previously.
involve a new research theme dealing with a range of This section covers future products from the new-gen-
matters from cloud particle genesis to rainfall decay in eration of geostationary satellites and related applica-
an integrated manner with a combination of radar and tions based on the key concepts detailed above.
satellite observations.
JMA plans to help mitigate the effects of natural 7.1 Optimal Cloud Analysis
disasters and resolve global environmental prob- To exploit the full potential of AHI multispec-
lems by maximizing the observational ability of tral observations, Optimal Cloud Analysis (OCA)
Himawari-8/9. To this end, the Agency established (Watts et al. 2011) will be introduced in collaboration
an application technology advisory panel in 2009 for with EUMETSAT. OCA involves use of the optimal
next-generation geostationary meteorological satel- estimation method (1D-Var) for the simultaneous
lites. The panel consists of professors and researchers retrieval of cloud and surface parameters (e.g., cloud
from Japanese universities and Japan Aerospace optical thickness, cloud effective radius, cloud top
Fig. 15. MSC website for provision of RGB composite images from Himawari-8 to users of SWFDP and SWFDDP
of WMO/CBS. [Available at http://www.data.jma.go.jp/mscweb/data/himawari/index.html.]
pressure, and surface temperature). Cloud microphys- cloud height data retrieved from OCA with refer-
ical parameters (e.g., effective radius) estimated from ence data such as satellite Cloud-Aerosol Lidar and
Himawari-8/9 AHI observations with high temporal Infrared Pathfinder Satellite Observation (CALIPSO)
resolution (an interval of 2.5 min around Japan) have products. A high level of consistency in the cloud
the potential to provide information on cloud genesis, height is observed from this figure.
evolution, and decay. JMA has run OCA with SEVIRI
data as a proxy for AHI data. Figure 16 compares
Fig. 16. Comparison between OCA cloud height data retrieved from SEVIRI and Cloud-Aerosol Lidar with
Orthogonal Polarization (CALIOP) data at 13 UTC on 1 October 2013 over Central Africa. (a) shows the height/
path cross section of CALIOP 523 nm total attenuated backscatter coefficients along a path (AB) of CALIPSO.
High-layer cloud heights (purple circles) and low-layer cloud heights (yellow circles) estimated using OCA are
superimposed onto the cross section. (b) shows a SEVIRI RGB composite image (red: 1.64 m; green: 0.81 m;
blue: 0.64 m). The red line from A to B corresponds to the path of CALIPSO shown in (a).
7.2 Atmospheric stability index product the Legacy Atmospheric Profile (LAP) product for
JMA plans to produce a stability index product GOES-R developed by NOAA/NESDIS (Lee et al.
for cloud-free conditions. This will be similar to 2014) and the Global Instability Index (GII) for MSG
developed by EUMETSAT (Koenig and Coning on visible images in the vicinity of a typhoon. The
2009). The product will provide atmospheric stability number of RS-AMVs is clearly larger. The ability
indices such as Convective Available Potential Energy of rapid-scan images to capture short-lived cloud
(CAPE), K-Index, and precipitable water values. The systems within typhoons supports their use for such
algorithm will involve the use of an optimal estima- compensation in data-sparse areas around typhoons.
tion method (e.g., 1D-Var) so that multispectral AHI JMA also conducted a study to compare ASCAT
information can be fully used to retrieve atmospheric wind and AMVs from rapid-scan visible imagery
profiles and calculate stability indices. The product covering the vicinity of a typhoon to validate the use
will support weather forecasters by enabling early of RS-AMVs in typhoon analysis. Figure 18 shows
detection of severe weather phenomena such as thun- the results of a comparison between RS-AMVs from
derstorms. visible imagery and ASCAT wind data from around
the vicinity of 2011s Typhoon Ma-on. Wind speed
7.3 Advanced convective cloud detection product comparison shows close correspondence, but the
JMA is currently developing a more advanced mean speed of ASCAT winds is about 0.8 times that
convective cloud retrieval product using the JMA of RS-AMVs up to 15 m s1. Wind direction differ-
Non-Hydrostatic Model (JMA-NHM) (Saito et al. ences are small compared to speed differences, but
2006) and the RSTAR radiative transfer model. The ASCAT winds are directed about 10 inward to the
evolution of local heavy rainfall is simulated using typhoon center with RS-AMVs as a reference. These
JMA-NHM, and successive satellite images are results indicate a close correlation between sea-sur-
then created at short time intervals by RSTAR as face winds and AMVs, as suggested by previous
the satellite simulator. The use of a numerical model work involving the use of sea-surface wind data
and a satellite simulator can provide deep insights from in-situ observations (e.g., Ohshima et al. 1991;
into the relationship between physical parameters Dunion and Velden 2002). Meanwhile, the wind speed
and their radiative signatures for convective cloud. of RS-AMVs is greater than that of ASCAT winds
If new information is found in this regard, data from in the high-speed region. This bias requires further
Himawari-8/9 may provide absolute evidence of rapid investigation. AMVs provided by Himawari-8/9
convection. The combination of Himawari-8/9 data on the basis of observations made every 10 or
and other observation data from radars, the light- 2.5 min are expected to be similar to MTSAT-1R
ning detection network, wind profilers, and upper RS-AMVs.
soundings is expected to improve the accuracy of
nowcasting for severe weather phenomena such as 7.5 Analysis of upper-tropospheric flows in typhoon
localized heavy rain, thunderstorms, and tornadoes. using AMVs
Using rapid-scan imageries from geostationary
7.4 AMVs for Nowcasting satellites obtained at intervals of less than 10 min,
AMVs derived from short-interval images tracking the clouds is enabled in typhoons where
are considered useful for capturing short-lived the direction of cloud motion abruptly changes and
phenomena such as rapidly developing cumulus active convections occur. The upper-tropospheric
cloud or rapidly deforming typhoon cloud systems. AMVs, which were retrieved from the tracks of
Advanced SCATterometer (ASCAT) surface wind upper-tropospheric clouds or water vapor patterns
data are currently utilized as a major ancillary from MTSAT-1R rapid-scan imageries, were applied
resource in JMAs typhoon analysis and, in particular, to analysis of wind field in typhoons. The anal-
for the estimation of gale force wind areas over data- yses of several typhoons revealed that both tangen-
sparse sea regions. However, as ASCAT coverage is tial and radial wind components of the AMVs,
temporally sparse (the ASCAT units on board Metop which correspond to cyclonic circulation and hori-
satellites observe the same points twice a day), data- zontal divergence in the upper troposphere near the
sparse regions need to be compensated for by other typhoon center, increased in the rapid intensification
observations. As mentioned in Section 6, MTSAT-1R stage of the typhoons. This result indicates that the
performed special rapid-scan (5-min interval) obser- AMVs reflected the intensification of the primary
vations around Japan during the daytime in summer. and secondary circulations of the typhoons during
Figure 17 shows an example of comparison between their intensification. It is expected that studies using
MTSAT-2 hourly AMVs and MTSAT-1R Rapid- upper-tropospheric AMVs from Himawari-8/9 for
Scan AMVs obtained every 5 min (RS-AMVs) based typhoons will contribute to an understanding of
Fig. 17. MTSAT-2 hourly AMVs (a, blue barbs) and MTSAT-1R rapid-scan AMVs (b, yellow barbs) based on
visible images in the vicinity of Typhoon Ma-on at 00 UTC on 23 July 2011
Fig. 18. Scatter plots of MTSAT-1R rapid scan wind from visible imagery and ASCAT wind data. (a) shows wind
speed, and (b) shows wind direction. The comparison period is from 17 to 23 July 2011, which is approximately
when Typhoon Ma-on was within MTSAT-1Rs rapid-scan area.
typhoon intensification processes and the diagnosis of data from these new observing platforms will be two
typhoon intensity change. orders of magnitude greater than before, the effective
use of such big data in very-short-range NWP is a big
7.6 Solar radiation estimation for photovoltaic challenge and may only be possible with a 10-petaflop
power system such as the K computer.
Expectations for the widespread utilization of
renewable energy forms such as photovoltaic power 7.8 Targeted observation
have been high in recent years, especially since the As detailed in Section 4, the AHI imager units on
Great East Japan Earthquake of 2011. Photovol- board Himawari-8/9 have the capability for flex-
taic power has significant potential to contribute to ible observation of the Target Area every 2.5 min
Japans electric power supply, but is usually subject to and Landmark Areas every 30 s to enable targeted
variations in solar radiation. Accordingly, the ability monitoring. Such targeted observation is an enhance-
to estimate and predict solar radiation accurately is ment of the operational observation network to be
of great importance. Geostationary meteorological assimilated into NWP. Additional observations will
satellites are useful tools in this regard because they be performed in sensitive regions as estimated from
are used to observe large areas of the earth with high ensemble forecasts, where these additional obser-
temporal frequency and moderate spatial resolu- vation data will be effective in improving predic-
tion. In this area, Takenaka et al. (2011) developed tion results. These approaches have been imple-
a new algorithm for solar radiation estimation from mented in a variety of field experiments, including
MTSAT-1R observation data using a neural network The Observing System Research and Predictability
based on radiative transfer in near-real-time. In Experiment (THORPEX) (Majumder et al. 2011),
this system, cloud properties are retrieved from the of which the THORPEX-Pacific Asian Regional
Comprehensive Analysis Program for Cloud Optical Campaign (T-PARC) was a sub-program. JMA began
Measurement (CAPCOM) (Nakajima and Nakajima contributing to this experiment in 2008 and executed
1995; Kawamoto et al. 2001). Plans are currently special observations such as dropsonde observation
being made to apply the huge observational ability of in collaboration with Germanys Falcon research
Himawari-8/9, which have high temporal frequency, aircraft, enhanced upper sounding using two research
high spatial resolution, and 16 bands, to the precise vessels and four automatic upper-sounding stations,
estimation of solar radiation. and MTSAT-2 rapid-scan operations (Nakazawa et al.
2010). The effectiveness of assimilating AMVs from
7.7 Super-rapid assimilation for super-rapid scan MTSAT-2 rapid scan observation data in T-PARC for
data from Himawari-8/9 NWP was validated by Yamashita (2010) and Wu
Heavy rainfall concentrated in very small areas has et al. (2014). Flexible observation by Himawari-8/9
caused major disasters in Japan in recent years. Such offers significant opportunities for operational
localized downpours are usually brought by convec- targeted observation. As a first step, experimental
tive clouds whose typical lifecycle is less than an application of these AHI functions is expected with
hour. In 2013, Japans RIKEN Advanced Institute sensitivity analysis based on ensemble prediction
for Computational Science began using the Japanese toward the goal of NWP innovation.
10-petaflop K computer in its Big Data Assimilation
8.Conclusions
science program for super-rapid 30-s cycles of an
ensemble Kalman filter to forecast such local severe As described in this paper, new-generation geosta-
weather at a 100-m resolution with a 30-min lead tionary satellite platforms such as Himawari-8/9
time. This forecast system is 120 times more rapid carry optical sensors with significantly higher radio-
than the current hourly updated ones and will help metric, spectral, and spatial resolution than those
prepare for sudden local torrential rainfalls that may previously available in geostationary orbit. These
cause flash flood and river outflow only within 1020 advantages, when combined with fast revisit times
min. In this work, observational data from Landmark of 10 min or less, will provide revolutions of capa-
Areas obtained every 30 s by Himawari-8 will also be bility for the identification and tracking of rapidly
assimilated with other sub-minute observation data changing weather phenomena. They will also
such as phased array weather radar information. These support the development of new surface and atmo-
data may be frequent enough to capture the nearly spheric monitoring/mapping applications in which
linear evolution of convection. As the volume of high temporal frequency and near real-time image
supply are essential. From another point of view, the images every 30 min (left) and Himawari-8 Japan
data set of AHI, which is the first 16 bands imager in Area images every 2.5 min at 11:0019:30 UTC on 26
geostationary orbit, serves as test data for the appli- January 2015.
cations ground segment of other new-generation satel-
Acknowledgments
lite imagers such as GOES-R/ABI, GEOstationary
KOrea Multi-Purpose SATellite-2A (GEO-KOMP- This paper has benefited from the contributions of
SAT-2A)/Advanced Meteorological Imager (AMI), numerous JMA staff. The authors thank everybody
MTG/Flexible Combined Imager (FCI), and so involved for their contribution. Thanks also go to the
on. JMA processes raw data from Himawari-8/9 members of JMAs Advisory Panel on Application
with quality control and creates HSD, imagery, and Technology for Next-generation Geostationary Mete-
physical products for its own use in NWP, severe- orological Satellites and related working groups. Dr.
weather watching, and environmental monitoring. Teruyuki Nakajima, who is a head of JAXA/Earth
The Agency also provides Level-1b data to external Observation Research Center (EORC), deserves
users via multiple channels including the Himawari- special thanks for his leadership as chairman of the
Cloud service and the HimawariCast service. advisory panel. Appreciation is also due to Prof.
Himawari-8/9 provides Full Disk observation data Eiichi Nakakita of Kyoto University and Associate
every 10 min by default. As such high-frequency Full Prof. Hirohiko Masunaga of Nagoya University for
Disk scanning is expected to become a world standard their fruitful input in discussions on collaborative
for emergency support in the near future, there is a research. The authors would also like to thank Prof.
need to develop and popularize the usage of high-fre- Takashi Y. Nakajima of Tokai University (Principal
quency data for disaster risk reduction in collabo- Investigator in CREST/TEEDDA) and Dr. Takemasa
ration with RSMCs and VAACs. It is also important Miyoshi of RIKEN AICS (Principal Investigator in
to promote Himawari-8/9 data utilization in the East CREST/BDA) for sharing their insights. Thanks also
Asia and Western Pacific regions within the frame- go to the Open Clustered Libraries for Atmospheric
work of CGMS. Wide-ranging application in various Science and Transfer of Radiation (OpenCLASTR)
aspects of meteorological, hydrological, and climate project for allowing the use of RSTAR package radia-
operation by countries in these areas is expected to tive transfer code in this research.
create a variety of new problems that will need to be
Appendix: List of acronyms
addressed for more effective usage of Himawari-8/9
data. In this regard, JMA remains committed to its ABI Advanced Baseline Imager
ongoing global contribution as a satellite operator and ACC ACCelerometer
an NMHS. AHI Advanced Himawari Imager
There is significant potential for the application of AIRS Atmospheric InfraRed Sounder
Himawari-8/9 data to various areas outside the mete- AMI Advanced Meteorological Imager
orological field, such as in other earth science fields AMV Atmospheric Motion Vector
and ecological/economic fields. To extract fruitful AORI Atmosphere and Ocean Research
results from these data, JMA is dedicated to its role Institute
in joint work with operators in the science commu- ARS Angular Rate Sensor
nity. Collaborative research agreements on satellite ASCAT Advanced SCATterometer
data quality control and application have already been BUFR Binary Universal Form for data
established between universities in Japan and JMA. Representation
Another example is JMAs engagement in collabo- CALIOP Cloud-Aerosol Lidar with Orthogonal
ration on joint international development toward the Polarization
creation of common scientific software for seamless CALIPSO Cloud-Aerosol Lidar and Infrared
determination of weather parameters by all meteoro- Pathfinder Satellite Observation
logical satellite operators. JMA will continue these CAPCOM Comprehensive Analysis Program for
domestic and international collaboration efforts in Cloud Optical Measurement
order to further develop satellite data applications. CAPE Convective Available Potential
Energy
Supplement
CBS Commission for Basic Systems
Supplement 1 shows infrared (IR1 or band #13) CGMS Coordination Group for Meteorolog-
animations from MTSAT-2 Northern Hemisphere ical Satellites
CSR Clear Sky Radiance LAP Legacy Atmospheric Profile

DCP Data Collection Platform LNB Low Noise Block converter
DCS Data Collection System LOS Line Of Sight
DVB-S2 Digital Video Broadcasting-Satellite- LRIT Low Rate Information Transmission
Second Generation MBCC Midnight Blackbody Calibration
ECMWF European Centre for Medium-Range Correction
Weather Forecasts MODIS MODerate resolution Imaging
ECV Essential Climate Variable Spectroradiometer
EMU Environmental Monitoring Unit MSC Meteorological Satellite Center
EORC Earth Observation Research Center MSG Meteosat Second Generation
ERA ECMWF Re-Analysis MTG Meteosat Third Generation
EUMETSAT EUropean organisation for MTSAT Multi-functional Transport SATellite
the exploitation of METeorological NASA National Aeronautics and Space
SATellites Administration
FCI Flexible Combined Imager NESDIS National Environmental Satellite,
GARP Global Atmospheric Research Data, and Information Service
Program NetCDF Network Common Data Form
GII Global Instability Index NHM Non-Hydrostatic Model
GISC Global Information System Centre NICT National Institute of Information and
GMS Geostationary Meteorological Satellite Communications Technology
GOES Geostationary Operational Environ- NIR Near-InfraRed
mental Satellite NMHS National Meteorological and
GEO-KOMPSAT-2A Hydrological Service
GEOstationary KOrea Multi-Purpose NOAA National Oceanic and Atmospheric
SATellite-2A Administration
GOS Global Observing System NWP Numerical Weather Prediction
GPCP Global Precipitation Climatology OCA Optimal Cloud Analysis
Project OpenCLASTR Open Clustered Libraries for
GSICS Global Space-based Inter-Calibration Atmospheric Science and Transfer of
System Radiation
GSM Global Spectral Model PFI Private Finance Initiative
GTS Global Telecommunication System PNG Portable Network Graphics
HOPE Himawari OPeration Enterprise QI Quality Indicator
corporation RDCA Rapidly Developing Cumulus Area
HRIT High Rate Information Transmission RGB Red, Green and Blue
HSD Himawari Standard Data RS Rapid-Scan
HSF Himawari Standard Format RSMC Regional Specialized Meteorological
IASI Infrared Atmospheric Sounding Centre
Interferometer RSTAR System for Transfer of Atmospheric
ICSU International Council of Scientific Radiation
Unions SATAID SATellite Animation and Interactive
ICT Internal Calibration Target Diagnosis
INR Image Navigation and Registration SCOPE-CM Sustained, Co-Ordinated Processing
IR InfraRed of Environmental satellite data for
IRU Inertia Reference Unit Climate Monitoring
ISCCP International Satellite Cloud SCOPE-NWC Sustained, Co-Ordinated Processing
Climatology Project of Environmental satellite data for
JAXA Japan Aerospace Exploration Agency NoWCasting
JDDS JMA Data Dissemination System SCT Solar Calibration Target
JMA Japan Meteorological Agency SEDA Space Environment Data Acquisition
JPEG Joint Photographic Experts Group SEVIRI Spinning Enhanced Visible and
JRA-55 Japanese 55-year Re-Analysis Infrared Imager
SNR Signal-Noise Ratio 92, 467475.

SRF Spectral Response Function Goodman, S. J., J. Gurka, M. DeMaria, T. J. Schmit, A.
STT STar Tracker Mostek, G. Jedlovec, C. Siewert, W. Feltz, J. Gerth,
SWFDDP Severe Weather Forecasting and R. Brummer, S. Miller, B. Reed, and R. R. Reynolds,
Disaster risk reduction Demonstration 2012: The GOES-R Proving Ground: Accelerating
user readiness for the next-generation geostationary
Project
environmental satellite system. Bull. Amer. Meteor.
SWFDP Severe Weather Forecasting Soc., 93, 10291040.
Demonstration Project Gunshor, M. M., T. J. Schmit, W. P. Menzel, and D. C.
THORPEX The Observing System Research and Tobin, 2006: Intercalibration of the newest geosta-
Predictability Experiment tionary imagers via high spectral resolution AIRS
T-PARC THORPEX - Pacific Asian Regional data. Preprints, 14th Conf. on Satellite Meteorology
Campaign and Oceanography, Atlanta, USA, 6.136.13.
VAAC Volcanic Ash Advisory Center Heidinger, A., 2011: Algorithm theoretical basis document
VIS VISible ABI cloud height. NOAA NESDIS Center for Satellite
WIS WMO Information System Applications and Research. [Available at http://www.
WMO World Meteorological Organization goes-r.gov/resources/docs.html.]
WWW World Weather Watch Holmlund, K., 1998: The utilization of statistical proper-
ties of satellite-derived atmospheric motion vectors
References to derive quality indicators. Wea. Forecasting, 13,
10931104.
Bormann, N., K. Salonen, C. Peubey, T. McNally, and C.
Japan Meteorological Agency, 2013: Outline of the oper-
Lupu, 2012: An overview of the status of the opera-
ational numerical weather prediction at the Japan
tional assimilation of AMVs at ECMWF. Proc. 11th
Meteorological Agency. Appendix to WMO Technical
Int. Winds Workshop, Auckland, New Zealand, 2024.
Progress Report on the Global Data-processing and
Dee, D. P., S. M. Uppala, A. J. Simmons, P. Berrisford, P.
Forecasting System (GDPFS) and Numerical Weather
Poli, S. Kobayashi, U. Andrae, M. A. Balmaseda, G.
Prediction (NWP) Research. Japan Meteorological
Balsamo, P. Bauer, P. Bechtold, A. C. M. Beljaars,
Agency, Tokyo, Japan. [Available at http://www.jma.
L. van de Berg, J. Bidlot, N. Bormann, C. Delsol,
go.jp/jma/jma-eng/jma-center/nwp/outline2013-nwp/
R. Dragani, M. Fuentes, A. J. Geer, L. Haimberger,
index.htm.]
S. B. Healy, H. Hersbach, E. V. Hlm, L. Isaksen, P.
JMA/MSC, 2006: Summary of MSC system. MSC technical
Kllberg, M. Khler, M. Matricardi, A. P. McNally,
note, Special Issue, 180 pp (in Japanese).
B. M. Monge-Sanz, J.-J. Morcrette, B.-K. Park, C.
Johnson, R. X., and M. P. Weinreb, 1996: GOES-8 imager
Peubey, P. de Rosnay, C. Tavolato, J.-N. Thpaut, and
midnight effects and slope correction. Washwell, E.
F. Vitart, 2011: The ERA-Interim reanalysis: Config-
R. (ed.), Proc. SPIE2812, GOES-8 and Beyond, 2812,
uration and performance of the data assimilation
596607.
system. Quart. J. Roy. Meteor. Soc., 137, 553597.
Kawamoto, K., T. Nakajima, and T. Y. Nakajima, 2001: A
Dunion, J. P., and C. S. Velden, 2002: Application of
global determination of cloud microphysics with
surface-adjusted GOES low-level cloud-drift winds
AVHRR remote sensing. J. Climate, 14, 20542068.
in the environment of Atlantic tropical cyclones. Part
Kieffer, H. H., and T. C. Stone, 2005: The spectral irradi-
I: Methodology and validation. Mon. Wea. Rev., 130,
ance of the moon. Astron. J., 129, 28872901.
13331346.
Kishimoto, K., M. Sasaki, and M. Kunitsugu, 2013:
Dvorak, V. F., 1975: Tropical cyclone intensity analysis and
Cloud Grid Information Objective Dvorak Analysis
forecasting from satellite imagery. Mon. Wea. Rev.,
(CLOUD) at the RSMC Tokyo - Typhoon Center.
103, 420430.
Technical Review of RSMC Tokyo-Typhoon Center,
Dvorak, V. F., 1984: Tropical Cyclone Intensity Analysis
13, 115.
Using Satellite Data. NOAA Tech. Rep. NESDIS, 11,
Kobayashi, S., Y. Ota, Y. Harada, A. Ebita, M. Moriya, H.
US Department of Commerce, National Oceanic and
Onoda, K. Onogi, H. Kamahori, C. Kobayashi, H.
Atmospheric Administration, National Environmental
Endo, K. Miyaoka, and K. Takahashi, 2015: The
Satellite, Data, and Information Service, 47 pp.
JRA-55 Reanalysis: General specifications and basic
Goldberg, M., G. Ohring, J. Butler, C. Cao, R. Datla, D.
characteristics. J. Meteor. Soc. Japan, 93, 548.
Doelling, V. Grtner, T. Hewison, B. Iacovazzi, D.
Kodaira, N., N. Murayama, H. Yamashita, and T. Kohno,
Kim, T. Kurino, J. Lafeuille, P. Minnis, D. Renaut, J.
1978: On the Geostationary Meteorological Satellite,
Schmetz, D. Tobin, L. Wang, F. Weng, X. Wu, F. Yu,
GMS (Himawari). Tenki, 25, 245268 (in Japanese).
P. Zhang, and T. Zhu, 2011: The Global Space-Based
Koenig, M., and E. de Coning, 2009: The MSG global insta-
Inter-Calibration System. Bull. Amer. Meteor. Soc.,
bility indices product and its use as a nowcasting tool.
Wea. Forecasting, 24, 272285. Nakajima, T. Y., and T. Nakajima, 1995: Wide-area determi-
Kosaka, Y., A. Okuyama, H. Takenaka, and S. Fukuda, nation of cloud microphysical properties from NOAA
2012: Development and improvement of a vicar- AVHRR measurements for FIRE and ASTEX regions.
ious calibration technique for the visible channel of J. Atmos. Sci., 52, 40434059.
geostationary meteorological satellites. MSC tech- Nakajima, T. Y., K. Suzuki, and G. L. Stephens, 2010a:
nical note, 57, 3955 (in Japanese). Droplet growth in warm water clouds observed
Lattanzio, A., J. Schulz, J. Matthews, A. Okuyama, B. Theo- by the A-Train. Part I: Sensitivity analysis of the
dore, J. J. Bates, K. R. Knapp, Y. Kosaka, and L. MODIS-derived cloud droplet sizes. J. Atmos. Sci.,
Schller, 2013: Land Surface Albedo from Geosta- 67, 18841896.
tionary Satelites: A multiagency collaboration within Nakajima, T. Y., K. Suzuki, and G. L. Stephens, 2010b:
SCOPE-CM. Bull. Amer. Meteor. Soc., 94, 205214. Droplet growth in warm water clouds observed by the
Lee, Y., Z. Li, J. Li, and T. J. Schmit, 2014: Evaluation of A-Train. Part II: A multisensor view. J. Atmos. Sci.,
the GOES-R ABI LAP retrieval algorithm using the 67, 18971907.
GOES-13 sounder. J. Atmos. Oceanic Technol., 31, Nakazawa, T., K. Bessho, S. Hoshino, T. Komori, K.
319. Yamashita, Y. Ohta, and K. Sato, 2010: THORPEX
Majumdar, S. J., S. Aberson, C. Bishop, C. Cardinali, J. - Pacific Asian Regional Campaign (T-PARC). Tech-
Caughey, A. Doerenbecher, P. Gauthier, R. Gelaro, nical Review of RSMC Tokyo-Typhoon Center, 12,
T. Hamill, R. Langland, A. Lorenc, T. Nakazawa, F. 14.
Rabier, C. Reynolds, R. Saunders, Y. Song, Z. Toth, Ohshima, T., H. Uchida, T. Hamada, and S. Osano, 1991:
C. Velden, M. Weissmann, and C.-C. Wu, 2011: A comparison of GMS cloud motion winds with
Targeted observations for improving numerical ship-observed winds in typhoon vicinity. Geophys.
weather prediction: An overview. World Weather Mag., 44, 2736.
Research Programme / THORPEX Publication, 15, Okada, K., J. Heintzenberg, K. Kai, and Y. Qin, 2001: Shape
37 pp. of atmospheric mineral particles collected in three
Maki, T., A. Ebita, T. Ishimizu, K. Nagata, M. Ikegami, and Chinese arid-regions. J. Geophys. Res. 28, 3123
T. Sasaki, 2008: The daily 3D ozone produced by 3126.
the chemical transport model for JRA-25. Extended Okuyama, A., A. Andou, K. Date, K. Hosaka, N. Mori,
abstracts of the Third WCRP International Confer- H. Murata, T. Tabata, M. Takahashi, R. Yoshino,
ence on Reanalysis, Tokyo, Japan. [Available at http:// and K. Bessho, 2015: Preliminary validation of
wcrp.ipsl.jussieu.fr/Workshops/Reanalysis2008/ Himawari-8/AHI navigation and calibration. Proc.
abstract.html.] SPIE 9607, Earth Observing Systems XX, 96072E,
Miyamura, K., 2007: MTSAT-2 systems. MSC technical doi:10.1117/12.2188978.
note, 49, 5558 (in Japanese). Olander, T. L., and C. S. Velden, 2007: The advanced
Menzel, W. P., W. L. Smith, and T. R. Stewart, 1983: Dvorak technique: Continued development of an
Improved cloud motion wind vector and altitude objective scheme to estimate tropical cyclone inten-
assignment using VAS. J. Climate Appl. Meteor., 22, sity using geostationary infrared satellite imagery.
377384. Wea. Forecasting, 22, 287298.
Meteo-France, 2012: Algorithm theoretical basis document Oyama, R., 2010: Upgrade of atmospheric motion vector
for cloud products (CMa-PGE01v3.2, CT-PGE02 derivation algorithms at JMA/MSC. MSC Technical
v2.2 & CTTH-PGE03 v2.2). NWC SAF/MSG, 87 pp. Note, 50, 132.
[Available at http://www.nwcsaf.org/HD/MainNS. Pavolonis, M., and J. Sieglaff, 2012: GOES-R Advanced
jsp.] Baseline Imager (ABI) Algorithm Theoretical Basis
Mohr, T., 2014: Preparing the use of New generation geosta- Document For Volcanic Ash (Detection and Height).
tionary meteorological satellites. WMO Bull., 63, Version 3.0, NOAA NESDIS Center for Satellite
4244. Applications and Research. [Available at http://www.
Murata, H., M. Takahashi, and Y. Kosaka, 2015: VIS and IR goes-r.gov/products/baseline-volcanic-ash.html.]
bands of Himawari-8/AHI compatible with those of Prata, F., 2011: Volcanic information derived from satellite
MTSAT-2/Imager. MSC technical note, 60, 118. data. EUMETSAT volcanic ash and SO2 pilot project
Nakajima, T., and M. Tanaka, 1986: Matrix formulation report, EUMETSAT.
for the transfer of solar radiation in a plane-parallel Remer, L. A., Y. J. Kaufman, D. Tanr, S. Mattoo, D. A.
scattering atmosphere. J. Quant. Spectrosc. Radiat. Chu, J. V. Martins, R.-R. Li, C. Ichoku, R. C. Levy,
Transfer, 35, 1321. R. G. Kleidman, T. F. Eck, E. Vermote, and B. N.
Nakajima, T., and M. Tanaka, 1988: Algorithms for radia- Holben, 2005: The MODIS aerosol algorithm, prod-
tive intensity calculations in moderately thick atmo- ucts, and validation. J. Atmos. Sci., 62, 947973.
spheres using a truncation approximation. J. Quant. Saito, K., T. Fujita, Y. Yamada, J. Ishida, Y. Kumagai, K.
Spectrosc. Radiat. Transfer, 40, 5169. Aranami, S. Ohmori, R. Nagasawa, S. Kumagai, C.
Muroi, T. Kato, H. Eito, and Y. Yamazaki, 2006: The tion of Moderate resolution Imaging Spectroradiom-
operational JMA nonhydrostatic mesoscale model. eter on EOS Aqua. J. Geophys. Res., 111, D09S05,
Mon. Wea. Rev., 134, 12661298. doi:10.1029/2005JD006 095.
Schmetz, J., P. Pili, S. Tjemkes, D. Just, J. Kerkmann, Tsuchiyama, H., Y. Kurihara, and K. Masuda, 2013: Devel-
S. Rota, and A. Ratier, 2002: An introduction to opment of volcanic ash product for the next-genera-
Meteosat Second Generation (MSG). Bull. Amer. tion Japanese Geostationary Meteorological Satellite
Meteor. Soc., 83, 977992. Himawari-8. Proc. 2013 EUMETSAT Meteorological
Schmit, T. J., M. M. Gunshor, W. P. Menzel, J. J. Gurka, J. Satellite Conference, Vienna, Austria.
Li, and S. Bachmeier, 2005: Introducing the next-gen- Uesawa, D., 2009: Clear Sky Radiance (CSR) product from
eration Advanced Baseline Imager on GOES-R. Bull. MTSAT-1R. MSC technical note, 52, 3948.
Amer. Meteor. Soc., 86, 10791096. Velden, C. S., T. L. Olander, and R. M. Zehr, 1998: Devel-
Schmit, T. J., J. Li, J. Li, W. F. Feltz, J. J. Gurka, M. D. opment of an objective scheme to estimate tropical
Goldberg, and K. J. Schrab, 2008: The GOES-R cyclone intensity from digital geostationary satellite
Advanced Baseline Imager and the continuation of infrared imagery. Wea. Forecasting, 13, 172186.
current sounder products. J. Appl. Meteor. Climatol., Velden, C., B. Harper, F. Wells, J. L. Beven II, R. Zehr, T.
47, 26962711. Olander, M. Mayfield, C. C. Guard, M. Lander, R.
Shimoji, K., 2014: Motion tracking and cloud height assign- Edson, L. Avila, A. Burton, M. Turk, A. Kikuchi,
ment methods for Himawari-8 AMV. Proc. 12th Int. A. Christian, P. Caroff, and P. McCrone, 2006: The
Winds Workshop, University of Copenhagen, Danish Dvorak tropical cyclone intensity estimation tech-
Meteorological Institute, Copenhagen, Denmark, nique: A satellite-based method that has endured for
S2-07. over 30 years. Bull. Amer. Meteor. Soc., 87, 1195
Stamnes, K., S. C. Tsay, W. Wiscombe, and K. Jayaweera, 1210.
1988: Numerically stable algorithm for discrete-ordi- Watts, P. D., R. Bennartz, and F. Fell, 2011: Retrieval of
nate-method radiative transfer in multiple scattering two-layer cloud properties from multispectral obser-
and emitting layered media. Appl. Opt., 27, 2502 vations using optimal estimation. J. Geophys. Res.,
2509. 116, D16203, doi:10.1029/2011JD015883.
Stohl, A., A. J. Prata, S. Eckhardt, L. Clarisse, A. Durant, S. Weinreb, M., and D. Han, 2003: Implementation of
Henne, N. I. Kristiansen, A. Minikin, U. Schumann, Midnight Blackbody Calibration Correction (MBCC).
P. Seibert, K. Stebel, H. E. Thomas, T. Thorsteinsson, NOAA NESDIS Office of Satellite Operations. [Avail-
K. Trseth, and B. Weinzierl, 2011: Determination of able at http://www.ospo.noaa.gov/Operations/GOES/
time- and height-resolved volcanic ash emissions and calibration/mbcc_implemmentation.html.]
their use for quantitative ash dispersion modeling: WMO, and EUMETSAT, 2007: Final report of RGB
The 2010 Eyjafjallajkull eruption. Atmos. Chem. composite satellite imagery workshop. Boulder,
Phys., 11, 43334351. United States. 14 pp.
Sumida, Y., T. Imai, K. Mouri, and A. Sobajima, 2015: Wu, T.-C., H. Liu, S. J. Majumdar, C. S. Velden, and J. L.
Convective cloud information derivation algorithm Anderson, 2014: Influence of assimilating satel-
theoretical basis document. MSC technical note, lite-derived atmospheric motion vector observa-
submitted. tions on numerical analyses and forecasts of tropical
Szejwach, G., 1982: Determination of semi-transparent cyclone track and intensity. Mon. Wea. Rev., 142,
cirrus cloud temperature from infrared radiances: 4971.
Application to METEOSAT. J. Appl. Meteor., 21, Xiong, X., W. L. Barnes, K. Chiang, H. Erives, N. Che, J.
384393. Sun, A. T. Isaacman, and V. V. Salomonson, 2004:
Tahara, Y., and K. Kato, 2009: New spectral compensation Status of Aqua MODIS on-orbit calibration and char-
method for intercalibration using high spectral reso- acterization. Proc. of SPIE, 5570, 317327.
lution sounder. MSC technical note, 52, 137. Yamashita, K., 2010: Observing system experiment of
Takenaka, H., T. Y. Nakajima, A. Higurashi, A. Higuchi, T. MTSAT-2 rapid scan atmospheric motion vector
Takamura, R. T. Pinker, and T. Nakajima, 2011: Esti- for T-PARC 2008 using the JMA operational NWP
mation of solar radiation by neural network based system. Proc. 10th Int. Winds Workshop, JMA,
on radiative transfer. J. Geophys. Res., 116, D08215, Tokyo, Japan, S6-06.
doi:10.1029/2009JD013337. Yang, P., K. N. Liou, K. Wyser, and D. Mitchell, 2000:
Tobin, D. C., H. E. Revercomb, C. C. Moeller, and T. S. Parameterization of the scattering and absorption
Pagano, 2006: Use of Atmospheric Infrared Sounder properties of individual ice crystals. J. Geophys. Res.,
high-spectral resolution spectra to assess the calibra- 105, 46994718.

An Introduction To Himawari-8 and 9

Uploaded by

Copyright:

Available Formats

An Introduction To Himawari-8 and 9

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

An Introduction To Himawari-8 and 9

Uploaded by

Copyright:

Available Formats

Journal

April of the Meteorological Society of Japan, Vol.K.94,BESSHO

Kotaro BESSHO, Kenji DATE, Masahiro HAYASHI

Meteorological Satellite Center, Japan Meteorological Agency, Kiyose, Japan

Satellite Program Division, Japan Meteorological Agency, Chiyoda, Japan

Meteorological Satellite Center, Japan Meteorological Agency, Kiyose, Japan

Hidekazu INOUE, Yukihiro KUMAGAI, Takuya MIYAKAWA

Satellite Program Division, Japan Meteorological Agency, Chiyoda, Japan

Meteorological Satellite Center, Japan Meteorological Agency, Kiyose, Japan

Satellite Program Division, Japan Meteorological Agency, Chiyoda, Japan

Meteorological Satellite Center, Japan Meteorological Agency, Kiyose, Japan

Typhoon Research Department, Meteorological Research Institute, Tsukuba, Japan

Yukio SASAKI, Yoshio SHIMAZU

Satellite Program Division, Japan Meteorological Agency, Chiyoda, Japan

Kazuki SHIMOJI, Yasuhiko SUMIDA

Meteorological Satellite Center, Japan Meteorological Agency, Kiyose, Japan

Corresponding author: Kotaro Bessho, Data Processing

Masuo SUZUKI, Hidetaka TANIGUCHI,

Satellite Program Division, Japan Meteorological Agency, Chiyoda, Japan

Volcanology Division, Japan Meteorological Agency, Chiyoda, Japan

Meteorological Satellite Center, Japan Meteorological Agency, Kiyose, Japan

Hironobu YOKOTA, and Ryo YOSHIDA

Satellite Program Division, Japan Meteorological Agency, Chiyoda, Japan

(Manuscript received 3 April 2015, in final form 7 December 2015)

Himawari-8/9a new generation of Japanese geostationary meteorological satellitescarry state-of-the-art optical

Keywords geostationary meteorological satellite; Himawari; satellite meteorology

(one for visible (0.500.75 m) and one for infrared

ronmental Satellite, Data, and Information Service

Fig. 2. Himawari-8/9 schedule.

Table 1. Major specifications of Himawari-8/9.

Fig. 3. Overview of ground segment for Himawari-8/9.

Table 2. Himawari-8/AHI and MTSAT-2/Imager specifications.

vapor, volcanic ash, sea surface temperature, and

4.2 Spatial resolutions

on the website at http://www.jma.go.jp/jma/jma-eng/ 4.3 Observation areas and frequencies

Fig. 7. AHI observation areas and frequencies on a timeline of 10 min.

Fig. 8. Simulated images for Himawari-8s 16 AHI bands based on RSTAR.

Table 3. Radiative transfer calculation design.

Table 4. Data set distributed via the HimawariCloud service.

Table 5. Data set continuously distributed via JDDS.

For rapid scanning observation, imagery in NetCDF 5.2 HimawariCast service

Table 6. Data set distributed via the HimawariCast service.

Table 7. Data set distributed via GISC-Tokyos SATAID service .

6.8 Contribution to ISCCP and GPCP 6.9 Contribution to SCOPE-CM

past satellites and provides the results to SCOPE-CM. climate research.

CSR Clear Sky Radiance LAP Legacy Atmospheric Profile

SNR Signal-Noise Ratio 92, 467475.

You might also like