Insat3d Catalog

0
INSAT-3D DATA PRODUCTS CATALOG
This catalog describes the INSAT-3D payloads, processing system

and data products which are available from the INSAT-3D
Meteorological Data Processing System (IMDPS) at National Satellite
Meteorological Centre (NSMC), India Meteorological Department,
New Delhi. A data product implied here is the valuable information
retrieved from INSAT-3D Imager and Sounder Payload data. Sample
of each product along with a short description of the algorithms,
processing steps, frequency, accuracy, applications and availability
are described.
1
1. Introduction ................................................................................ 5
2. Spacecraft: .................................................................................. 6
3. INSAT-3D Payloads ...................................................................... 7

3.1 Imager ...................................................................................................................................... 7
3.2 Atmospheric Sounder ................................................................................................................ 9
3.3 Mode of operation of INSAT-3D, Imager and Sounder: ............................................................... 10
3.4 Data Relay Transponder (DRT): ................................................................................................. 10
3.5 Satellite Aided Search and Rescue (SAS & R) Transponder .......................................................... 11
4.INSAT-3D Meteorological Data Receiving Earth Station (IMDRES) .. 12

4.1 Earth Station & Baseband System: ............................................................................................ 12
4.2 Quick Look Processing & Archival System .................................................................................. 14
4.3 IMDRES Software Components ................................................................................................. 15
5. INSAT-3D METEOROLOGICAL DATA PROCESSING SYSTEM (IMDPS) 15

5.1 System Overview ..................................................................................................................... 16
5.2 Data Acquisition and Quick Look Display System (DAQLS)/ Data Reception(DR) System ............... 16
5.3 Data Processing System (DPS)................................................................................................... 18
5.4 IMDPS Data Products Generation Software (DPGS) ................................................................... 19
5.5 Data Dissemination Server ....................................................................................................... 23
5.6 Ancillary Data Processing System (ADPS) ................................................................................... 23
5.7 Database Management System ................................................................................................ 24
5.8 Product Monitoring and Management System (PMMS) and PSS ................................................. 24
5.9 Data Dissemination Software (DDS) .......................................................................................... 24
6. Geo-physical Product Generation................................................ 25
2
7. Geophysical (GP) parameter retrieval.......................................... 25
8. INSAT-3D Data Products Type and Format ................................... 27

8.1 Product File Naming Convention............................................................................................... 27
8.2 INSAT-3D Data Products Types ................................................................................................. 27
9. IMDPS Storage Infrastructure ..................................................... 31

9.1 Storage Area Network/Online SAN Storage.............................................................................. 31
10. Network Clients System ........................................................... 33
11. Satellite Image Display System (SIDS) ........................................ 33
12. INSAT-3D IMAGER SAMPLE DATA PRODUCTS ............................. 35

12.1 Imager FULL GLOBE Images .................................................................................................... 36
12.2 Imager Sector Images............................................................................................................. 40
13. INSAT-3D IMAGER GEO-PHYSICAL PARAMETERS ........................ 41

13. 1 Outgoing Long wave Radiation (OLR)...................................................................................... 42
13.2. Satellite Rainfall Estimation ................................................................................................... 44
13.3 Operational Implementation of High-Resolution Rainfall Measurements (Hydro-Estimator) ....... 47
13.4 Sea Surface Temperature (SST) .............................................................................................. 49
13.5 Cloud Motion Vectors (CMV) ................................................................................................. 52
13.6 Water Vapor Wind Vector (WVWV)........................................................................................ 56
13. 7 Upper Tropospheric Humidity (UTH) ...................................................................................... 60
13.8 Snow-Cover Mapping ............................................................................................................. 62
13.9 Snow Depth Estimation .......................................................................................................... 65
13.10 FIRE IDENTIFICATION........................................................................................................... 67
13.11 Smoke Identification (SI)....................................................................................................... 69
3
13.12 AEROSOL OPTICAL DEPTH (AOD) ........................................................................................... 72
13.13 FOG..................................................................................................................................... 74
13.14 Cloud Mask (CMK) .............................................................................................................. 77
14. SOUNDER ATMOSPHERIC PARAMETERS .................................... 79
15. Application parameters from Sounder derived products ............ 83

15.1 GEOPOTENTIAL HEIGHT.......................................................................................................... 84
15.2 Total and Layer Precipitable Water ......................................................................................... 86
15.3 LIFTED INDEX ......................................................................................................................... 90
15.4. DRY MICROBURST INDEX ....................................................................................................... 92
15.5 MAXIMUM VERTICAL THETA-E DIFFERENTIAL .......................................................................... 94
15.6 WIND INDEX .......................................................................................................................... 95
16. DATA AVAILABILITY.................................................................. 97
4
1. Introduction
INSAT-3D:India launched an exclusive meteorological satellite on 26th July, 2013
from French ,Guyana using ARIANE rocket. The INSAT-3D is an exclusive
meteorological satellite with the following mission objectives:
 To monitor earth’s surface, carryout oceanic observations and its

environment in various spectral channels of meteorological importance.
 To provide the vertical profile of temperature and humidity parameters of
the atmosphere.
 To provide the data collection and data dissemination capabilities from the
Data Collection platforms (DCPs).
 To provide the satellite aided search and rescue services.
To accomplish the mission objectives, INSAT-3D has following payloads on board

the spacecraft:
• Six channel imager

• Nineteen channel sounder
• Data Relay Transponder(DRT)
• Satellite aided Search and Rescue(S&SR) System
Meteorological payloads are state-of-art and have significant technological

improvement in sensor capabilities and higher resolution compared to earlier
INSAT missions. The Imaging System of INSAT-3D has the following significant
improvements over that of KALPANA and INSAT-3A:
• Imaging in Middle Infrared band to provide night time pictures of low

clouds and fog.
• Imaging in two Thermal Infrared bands for estimation of Sea Surface
Temperature (SST) with better accuracy.
• Higher Spatial Resolution in the Visible and Thermal Infrared band.
INSAT-3D adds a new dimension to weather monitoring through its atmospheric

Sounding System, which provides vertical profiles of temperature (40 levels from
surface to ~ 70 km), humidity (21 levels from surface to ~ 15 km) and integrated
ozone from surface to top of the atmosphere. INSAT-3D provides continuity to
earlier missions and further augment the capability to provide various
meteorological as well as search and rescue services.
It is also proposed to establish a land based CAL/VAL site for INSAT-3D satellite
for Visible and SWIR channel over Jaisalmer, Rajasthan, which would be suitable
for calibrating and validating the radiometric gain of an in-flight satellite imaging
5
optical sensor.A joint-campaign to this effect has already been carried out
atJaisalmer site with NRSC,RRSC,SAC,(ISRO), IITM and NPL scientists during
14-23 December 2013. The outcome of the observation results will finalize the
CAL/VAL site.
2. Spacecraft:
The INSAT-3D islocated at 82Degrees East. It is a momentum-biased 3-axis body
stabilized geostationary spacecraft, using star trackers for precise pointing
control. It has well proven I-2K buswhich is built around the imaging payload
requirement to maintain mass and volume to a minimum, to meet the payload
thermal control requirements, locating the electronics from the signal to noise
ratio consideration etc. The spacecraft has dry mass of 958.5 kg. The nominal
design life is 7.7 years. One sided solar array of 1.9mx2.8m panel size has been
considered on south surface to meet the power requirements. Also it avoids the
thermal loads on payloads providing a better thermal regime for payloads. The
payload Electro Optics Modules (EOMs) are on the payload deck (on the North-
East corner). All the payload electronic packages are mounted on the underside
of North equipment panel.
INSAT-3D configuration details

6
Salient Features
MISSION Meteorological and Search & Rescue Services
ORBIT Geostationary
LOCATION 82 deg E Longitude
MASS AT LIFT- 2060 Kg

OFF
DIMENSIONS 2.4m x 1.6m x 1.5m
MISSION LIFE 7.7 years
PAYLOADS Imager, Sounder, Data Relay Transponder and Satellite

Aided Search & Rescue (SAS&R) transponder
3. INSAT-3D Payloads
3.1 Imager
The INSAT-3D Imager is an improved design of VHRR/2 (Very High Resolution

Radiometer) heritage instrument flown on the Kalpana-1 and INSAT-3A missions.
INSAT-3D carries a multi-spectral Imager (optical radiometer) capable of
generating the images of the earth in six wavelength bands significant for
meteorological observations, namely, visible, shortwave infrared, middle infrared,
water vapor and two bands in thermal infrared regions, offering an improved 1
km resolution in the visible band for the monitoring of mesoscale phenomena and
severe local storms. Imager specification is as follows.
7
Imager Specifications
Spectral Wave length Ground Quantization IGFOV

Band m Resolution bits  rad
VIS 0.55 – 0.75 1 Km 10 28
SWIR 1.55-1.70 1 Km 10 28
MIR 3.80-4.00 4 Km 10 112
WVP 6.50-7.10 8 Km 10 224
TIR 1 10.3-11.3 4 Km 10 112
TIR 2 11.5 – 12.5 4 KM 10 112
The two new SWIR and MWIR bands with a resolution of 1 km and 4 km,
respectively, enables better land-cloud discrimination and detection of surface
features like snow. One more significant improvement is the split-band TIR
channel with two separate windows in 10.2-11.2 and 11.5-12.5 μm regions with
a 4 km resolution.
This new element helps in the extraction of sea surface temperature over the
Indian region with far greater accuracy since the dual-window algorithm can be
applied to eliminate the atmospheric attenuation effects. The 1 km resolution of
the visible channel and 4 km resolution of the thermal IR channels improves the
accuracy of the derived products like outgoing long wave radiation and cloud
motion vectors.The Imager generates images of the earth disk from geostationary
altitude of 36,000 km every 26 minutes and provide information on various
parameters, namely, outgoing long-wave radiation, quantitative precipitation
estimation, sea surface temperature, snow cover, cloud motion winds, etc.
The salient features of INAST-3D Imager are as follows:

1. Blackbody calibration sequence is modified as compared to VHRR of earlier
satellites.
2. Two flexible mode of operation:
 Full frame mode scans 18 degree EW x18 degree NS covering the
entire Earth disc.
8
 Program mode covering 18 degree in EW direction NS coverage can
be defined in terms of number of lines to be scanned.
3. High Resolution mode: in the Fast Scan direction IFOVs are over sampled
by 1.75 times.
3.2 Atmospheric Sounder
INSAT-3D carries a newly developed 19 channel sounder, which is the first such
payload to be flown on an ISRO satellite mission. The Sounder has eighteen
narrow spectral channels in shortwave infrared, middle infrared and long wave
infrared regions and one channel in the visible region. The ground resolution at
nadir is nominally 10x10km for all nineteen channels. The specification of
Sounder is as follws.
Sounder Specifications
Channel Spectral Range Resolution

microns
VISIBLE 0.67 – 0.72 10X 10 kms.
SWIR 3.67 – 4.59 10X 10 kms
MIR 6.38 – 11.33 10X 10 kms
LWIR 11.66 – 14.85 10X 10 kms
It provides information on the vertical profiles of temperature, humidity and

integrated ozone. These profiles are available for a selected region over Indian
landmass every one hour and for the entire Indian Ocean Region every six hours.
The salient features of INSAT-3D sounder design are as follows:

1. Blackbody calibration sequence is modified as compared to VHRR of earlier
satellites.
2. In order to improve noise performance, facility to collect two or four
samples (0.2 sec or 0.4 sec step & dwell time) of the same area also which
can then be processed on ground. This will increase the sounding time
proportionally.
3. A biannual rotation of yaw by 180 degree has been introduced to reduce
the cooler patch temperature. This is to be taken care during processing.
9
3.3 Mode of operation of INSAT-3D, Imager and Sounder:
There are three modes of operation of INSAT-3D satellite for Imager & Sounder
payloads:
Mode of Operation Time of coverage Coverage Area

Full frame mode 26 minutes 18x18 degrees
ProgrammedNormal 23 minutes 14x18degrees
scan mode
Programmed Sector 6 minutes 4 degrees in NS & 18
scan mode degrees in EW
Sounder Scan Schedule (6 Hour cycle):

00:00-00.18Z : Region A1 00:20-00.52Z : Region A2
05:00-05.52Z: Region B
A 1
A 2
A 1 : 2 x 5 ( 1 8 M in )
A 2 : 3 x 6 ( 3 3 M in )
B : 4 x 7 ( 5 1 M in )
3.4 Data Relay Transponder (DRT):
Data Relay Transponder (DRT) on-board INSAT-3D is very useful for receiving
meteorological, hydrological and oceanographic data from remote, uninhabited
locations over the coverage area from Data Collection Platforms (DCPs) like
Automatic Weather Station (AWS), Automatic Rain Gauge (ARG) and Agro Met
Stations (AMS). The data is relayed back for down linking in extended C-Band
10
INSAT Data Collection Platform Elements of Data Collection Platform
For extreme weather related disasters such as cyclone, floods and drought, real
time observations of the associated parameters with appropriate network density
is very important. Satellite enabled Data Collection Platforms provide a unique
solution for gathering meteorological data from all over the country including
remote and inaccessible places. India Meteorological Department (IMD) and ISRO
have established more than 1800 Data Collection Platforms.INSAT-3D provides
continuity of service of DRT which is currently carried by KALPANA-1 and INSAT-
3A.
3.5 Satellite Aided Search and Rescue (SAS & R) Transponder:
INSAT-3D is equipped with a Search and Rescue payload (operating in 406 MHz)
that picks up and relays the alert signals originating from the distress beacons of
maritime, aviation and land based users to the Indian Mission Control Centre
(INMCC) located at ISRO Telemetry, Tracking and Command Network (ISTRAC),
Bangalore. The major users of Satellite Aided Search and Rescue service in India
are the Indian Coast Guard, Airports Authority of India (AAI), Directorate General
of Shipping, Defence Services and fishermen. The Indian service region includes a
large part of the Indian Ocean region covering India, Bangladesh, Bhutan,
Maldives, Nepal, Seychelles, Sri Lanka and Tanzania for rendering distress alert
services. INSAT-3D joins INSAT-3A to provide operational Search and Rescue
service
11
4.INSAT-3D Meteorological Data Receiving Earth Station
(IMDRES)
Master Control Facility (MCF) Hasan, ISRO on behalf of ANTRIX has established,
the ‘INSAT-3D Meteorological Data Receiving Earth Station (IMDRES)’ comprising
of the ground reception infrastructure, Quick Look Processing & Archival system
and Monitoring and Controlling system at IMD premises in New Delhi.
4.1 Earth Station & Baseband System:

The station incorporates 7.2m shaped Cassegrain antenna with Ext. C-band,
Dual linearly polarized, receive feed system & operation in the Ext. C-band
frequency range 4.5 to 4.8 GHz in receive band. The antenna pedestal is an
elevation-over-azimuth type, providing adequate steer ability to look at
geostationary satellites. Azimuth and Elevation axis are motorized. The antenna
is integrated with Antenna control system with various built-in modes for
tracking operation. The system comprises of Downlink chains to receive Imager,
Sounder and AWS Data from INSAT-3D. Sufficient redundancy is incorporated to
handle any failures and to enable IMD to receive the uninterrupted data from
INSAT-3D.
12
IMDRES comprises of six RF chains. They are designated as two mutually
redundant RF Chains for each Sounder & Imager, an AWS chain and a spare
chain. It also comprises of adequate RF and IF instruments for necessary
measurements/ monitoring and two mutually redundant Quick Look Processing
& Archival systems configured to the RF chains through Interface Boxes.
13
4.2 Quick Look Processing & Archival System
The Quick Look Processing & Archival System is a vital element at IMD for day-
to-day operations of satellite meteorology. It is expected to cater to the following
specific requirements of IMD.
• A quick information about the downlink signal LOCK Status.
• A quick Look Image Display in real-time to ensure proper reception of

image data.
• Automated Image Data Archival into Network Attached

Storage(NAS),which serves as a backup data repository for further
processing at IMD as well as an emergency data source for MCF.
• A Quick reference to few informatory parameter status/values from the

onboard imaging payloads in real-time.
14
4.3 IMDRES Software Components
The software enables the operation of system in two modes:
• User Mode: For routine operations by the end user (operator). Since the
software is expected to provide mainly the quick look display, most of the
operations are automated with minimum user interaction.
• System Administrator Mode: For the system configuration management,

routine
housekeeping, automation control, etc by the system administrator.
Acquisition Module of the software is responsible for the data acquisition. Signal
lock is shown by GREEN indicator and the unlock status is shown by RED. The
Application Module of the software provides the following.
• Data Archival: Automatic full data archival into NAS as well as the
average data archival for each imaging operation.
• Image Display: For quick look image displays of all channels.
• Page Display: For referring the informatory parameter display.
• Real Time Graph: For a quick glance of the trends of few parameters.
M&C System provides access to the equipments residing in the earth station.The
system is designed based on client-server concept. The Software uses TCP/IP
protocol. The M&C server software executing at earth station computer
communicates with all equipments through RS-232 cables. The client software
executing in client computer (located in IMD building) communicates with the
server through network and collects all the information about the equipments of
the earth station.
5.INSAT-3D METEOROLOGICAL DATA PROCESSING SYSTEM

(IMDPS)
Indian Space Research Organization (ISRO) provides end-to-end solution for
reception and processing of INSAT-3D data and derivation of meteorological
parameters. An indigenously designed and developed INSAT-3D Meteorological
Data Processing System (IMDPS) has been installed and commissioned at IMD,
New Delhi with a Mirror Site at Space Applications Centre, Bopal, Ahmedabad.
IMDPS caters to the processing of all data transmitted by the Imager, Sounder
and DRT payloads.IMDPS also simultaneously processes data of INSAT-3A and
Kalpana-1 satellites. The data archival and dissemination is through IMD, New
Delhi.The IMDPS system acquires raw data from serial data streams, processes
the data and generates various quantitative products from the processed data for
15
operational utilization by various users. Additionally, the system is capable of
processing, ingesting and analyzing Automatic Weather Station (AWS) and Global
Telecommunication System (GTS) data.
5.1 System Overview
The Requirements of INSAT-3D Meteorological Data Processing system (IMDPS)

are realized using specialized software and hardware. Following are the major
sub-systems for processing of INSAT Meteorological Data. The main architecture
of IMDPS system consists of Data Reception system, (DRS) Data Processing
System (DPS), Data Storage System (DSS), Network Clients System (NCS) and
Satellite Images Display System (SIDS)
5.2 Data Acquisition and Quick Look Display System (DAQLS)/ Data
Reception(DR) System
The DAQLS and DR system is the front-end system for Data Acquisition and
processing chain to capture, archive, display and transferring the RAW data and
Meta data to the DP System for subsequent Processing. Each DR system receives
the base band serial data stream of the satellite sensor from the corresponding
RF-IF segment. Every sensor (Imager, Sounder, VHRR and CCD) base band data
streams are handled independently. DR systems for each sensor stream are
implemented in redundant configuration. The major functions are
• Frame synchronization,
• Frame and Format verification,
• Data Acquisition,
• Raw data archival
• Processed Quick Look Display (P-QLD), HK display and logger,
• Data transfer to DP, Raw data Replay suits,
• Online BER measurement
16
The main functions like Data Acquisition and Raw data archival are implemented
in these systems. The DR systems are connected to Data Processing System
through TCP/IP network. Each DR system transfers a processed raw sensor
image to the connected Data Processing computer immediately after the
compilation of Image.
The Hardware elements of the DR systems are,
a. Bit synchronizer Units

b. Computer Interface Units (CIUs)
c. Data Reception Servers (DR Servers)
d. Raw Archival Disks
e. Display work station
f. LTO drive and media
Bit Sync. Unit receives the base band serial stream from the Demodulator, and
provides the Serial Randomized NRZ-L data along with synchronized clocks. It
generates TTL serial data along with synchronized 0 deg. and 90 deg. clocks. It
also performs the functions of
 Code conversion from NRZ-S to NRZ-L (for VHRR, SOUNDER)

 Clock hold in case of short period breaks in the data stream.
 Track and hold the output data to the specified frequency in case of
disturbances in demodulator output
The CIU unit(s) receives the Serial Base band data with clock from the Bit
Synchronizer(s). They perform the functions of Frame Synchronization, DE
randomization, and formatting the data.
Real time Data Acquisition, Disk Archival and DP image data transfer, P-QLD,
RAW Displays are implemented on the DR Server. It also includes a Telemetry
logger and a real time display of some telemetry items. The software is
implemented such that Data Ingest along with all applications can be operated at
the same time on the single Server. The DR systems are configured on a dual
CPU server with OVMS operating system. Each Server is capable to process two
independent base band chains, with all the required DR functions.
Raw Archival Disks and LTO-3 drives are directly connected with DR servers
through SCSI interface for raw data backup. Printers are connected with LAN
switch. Additional Display workstations to support Online QLD and HK Displays
are connected to the Servers through network.
17
DR software suites run on the DR Servers. Two independent such suites (one
each for the base band chain configured on the server) are normally operational
on each server (VHRR/CCD/Sounder/Imager).
The DR software consists of suites for
• Data Acquisition and CIU control and Ingest, with automatic transfer to
“DP” computers.
• Real time data acquisition and ingest with online status updates.
• Band wise Quick Look Display (partially processed) on Console and
networked display workstations.
• Raw Data Archival, with secondary archival on LTO.
• Telemetry logger.
• Online processing of raw data stream for band separation, telemetry
stripping.
• Offline data transfer of raw Image data to “DP” computers DR software
suite implementation aresuch that they can be operated individually and
independently in a very flexible manner ( eg. P-QLD , TM processing,
Ingest are all independent)
The numbers of application tasks like P-QLDs, HK plots, HK displays are

automatically limited for a given Sensor data stream to avoid server resource
crunch. The main Data acquisition and Ingest suits module distributes the raw
data to all other application modules.
DR software for a given DR-Main and DR-Redundant chain operate

independently and exclusively with respect to one another. Two independent
such suites (of different Sensor data streams) can be operated on each server.
DR software is implemented using C++, Fortran, System services of O/S, X-
window GUI, on the OVMS server platform.
5.3 Data Processing System (DPS)
The Data Processing System (DPS) is main component of the IMDPS system. The
raw data from DRS is transferred to the Data Processing System via TCP/IP
network. The main function of DPS is to process the data received from DRS and
give the desired products. The DPS consists of 7 main severs having 5 DELL PE
6850 servers (INSAT-3D Imager, INSAT-3D Sounder, INSAT-3A CCD, INSAT-3A
VHRR and KALPANA-1 VHRR servers), a DELL PE 2950 server (Image Analysis
Server) and 2 HP xw6400 servers (Ancillary Data Processing System & CCD
application server). All the raw data from DRS is processed in these servers & the
desired products are obtained. After processing the data is transferred to an
external storage system for storing the data.
18
Meta Data
DR-Main
Database Server Meta Data
DP NAS
RAW Data
DR-Redundant
DP Processing DP NAS DDS

DP SAN
Products
PMMS
Web/SIDS/
Archive
5.4 IMDPS Data Products Generation Software (DPGS)
The main function of INSAT-3D data Products Generation software is to process

the raw data received from Data Acquisition and Quick Look System (DAQLS)
and generate predefined data products. The Processing level and supported
formats for each type of data products are given in subsequent paras.
DPGS performs the following functions:
Generates the ephemeris information for the given date, time of

acquisition.
Extracts different band images, space look data, black body data, AOCS,
Fast/Slow scan servo error information and the ancillary information ATM
(Analog and Digital Telemetry), DTM ( Digital telemetry) from the raw data
Corrects for the servo error
Radiometric corrections
Geometric corrections
Generates supported digital data product in HDF5
Generates various pre-defined Geo-physical parameter Products.
Dissemination of Products
19
Data Products Generation Chain
Level-1 Hdf Product Level-1 Product

Data Reception Generaton generation
Data Transfer to DGPSs Geometric corrction Level-2 Geophysical
Band Rediomatic correction

Level-3 Binned product
seperation,Ephemeris and Black body
generation
Extraction calibration
5.4.1INSAT-3D Data Products Corrections
5.4.1.1 Raw data Extraction:Image data along with telemetry (both analog and
digital), servo profile, attitude, space look and black body information are
extracted for each acquisition.
Servo Correction: The mirror position is commanded from ground. Suppose,

during one fast scan sweep the commanded angle is . However, due to various
reasons, the mirror might image at position +. In the next frame, at the same
position as above, the commanded position is again , but the mirror might
image at +’. This difference in the actual position of the mirror in two
successive frames when commanded to look at the same portion of the earth
gives rise to a break in features. A linear feature will look broken. The error  is
called servo error. If the image has to be corrected, this break in the features has
to be removed. This process is called servo correction. Unless this is done, even
the earth disk will not look a full circle, but rather like a circular gear wheel.
Servo error profile obtained from raw data is used to correct servo error.
Following corrections are done as part of servo correction:
Servo Correction
Stagger Correction
Oversampling Removal
20
Linearity Correction
5.4.1.2 Radiometric Corrections
The response of the different detectors in the array has to be normalized for
correct interpretation by different applications. This is done by generating a
Calibration Lookup table. Before launch, the calibration lookup table is generated
in the Laboratory. However, after launch, these values tend to change. The
change is detected using the Cold Space and Black body data available with every
acquisition. Using this cold space data and the on board temperature at the time
of acquisition, new calibration look up table is generated. The calibration lookup
table is provided with every product to the user.
Following corrections are done as part of radiometric corrections:
Detector normalization
Line loss correction
Black Body Calibration and BT (Brightness Temperature) Table
Generation
Radiometric Normalization
Line Loss Correction
Stripe Removal (if any)
5.4.1.3 Geometric Corrections
The aim of Geometric correction is to assign proper geo-

location (latitude/longitude) coordinates to the different pixels of the image data
and compute viewing angle for satellite and Sun. Following corrections are done.
Navigation (Computation of Geo-location using ephemeris, attitude and

imaging model)
Computation of Satellite/Sun Azimuth/Elevation
Yaw Flip Correction
 System level Geo referencing for small area of interest (AOI) defined by the
user– Sector Product. Following additional corrections are done for sector
products of Map projection resampling
Boundary file generation

Residual Attitude Computation
Image and Mirror Motion Compensation (IMC/MMC) have been incorporated

in INSAT-3D. However the effect of this will be reflected only in the mirror
position. This is available as part of Auxiliary data to Data Products System. The
DTM contains twenty samples of the fast scan mirror positions for every frame.
Polynomial fit through these twenty points and this curve is used to interpolate
21
the fast scan mirror positions at other locations. Similar interpolation is done to
calculate the slow scan mirror positions.
Another new element in INSAT-3D is the half yearly yaw flip– the satellite will be
rotated by 180 degrees once in six months. The DP software takes care of the yaw
flip effect. The information is available in the form of yaw angle and using this
product are generated in the 'upside' mode.
Imaging Mechanism
The payload consists of a scan mirror that rotates in the west to east (east to
west) direction over an angle of 9. This sweep covers a field of view of 18 on the
earth in the west to east (east to west) i.e Fast Scan direction. Each of these
sweeps is called a frame. This sweep is repeated times, by changing the angle of
the mirror in the north south direction. The starting angle for the slow scan can
be set at users' request, thus enabling coverage of different areas.
Navigation: Geolocation
The first step is to find the coordinates of the points imaged by the various
detector elements during the full slow/fast scan. The slow scan and fast scan
angles are obtained from the telemetry data coming along with the video data.
The satellite state vector is obtained by using an orbit determination algorithm.
The inputs for this are obtained from Mission Control. Since the satellite is
stationary and the field of view is large in both north-south and west-east
directions, sometimes, the satellite might look into cold space. This is taken care
in the model.
In one frame, the slow scan varies from say Φst to Φend, while the fast scan angle
varies fromst toend.
For
Φst ≤ Φ≤Φend, st≤ ≤end
Using the geometric model and modeling the earth as a spheroid, the point on the
ground imaged by the particular detector element is found out as (xg, yg, zg) in
the inertial frame of reference
Thus, the mapping
(detector_no,,) (, )
is established.
This map gives the geolocation information of the data acquired.
System level Geocoding
Geocoding is the process of projecting the satellite image data into a proper map
projection with a constant spatial resolution, and aligning the same to true
22
North, so that the data can be overlaid on a map. For this, the corner coordinates
are given apriori. These are converted into map projection coordinates in the
desired projection.
5.4.2Product Formatting and Dissemination
All products are generated in HDF5 (Hierarchical Data Format) format and
corresponding image chips are generated in JPEG format. Atmospheric Motion
vectors (AMVs) are generated in BUFR (Binary Universal Format
for Representation of Meteorological data) format also. All products are
disseminated through IMD web and GTS network.
5.5 Data Dissemination Server
Various Products are disseminated by Rule based Data Dissemination Software.

Based on requirement dissemination rule can be modified to include new
products for dissemination.
Few of the functionalities include
 Transfer of Imager to the web site.

 Conversion of products to GTS format and transmission to
Meteorological communications computer.
 Encoding of products and AWS data in WMO format and transmission
to Meteorological Communications Computers (MCC).
5.6 Ancillary Data Processing System (ADPS)

The Ancillary Data Processing system caters to the requirements of ingesting
data from other sources and conversion of the data to the required format for
archival of the same at the central data storage. This archived data is used by
Data Products/Parameter Retrieval system for processing.
23
5.7 Database Management System
The Database Management System is using ORACLE 10g as a backend Database
server and it contains Metadata of all processed products available on the central
store. This database also contains information about the permanent databases
such as GCP and boundary database, Ingested Auxiliary Data i.e. data from GTS
and AWS and other information, which are used during the Data Products
Generation and by Image Analysis software.
5.8 Product Monitoring and Management System (PMMS) and PSS
The Product Monitoring and Management Software (PMMS) have an interface

with the Database management system and Process Scheduling Server (PSS),
which runs on all configured Data processing systems.
The PMMS is capable of displaying the current processing status as well as the
status of products already processed on configured data processing system. The
GUI automatically updates the status of the process and displays the status by
using proper coloring scheme. This software also has capability to display all
sub-sampled data associated with the product for all acquisition for the operator-
selected date. Operator is also provided capability to display images associated
with the final product in full resolution as part of product visualization.
The Software requirements can broadly be classified under following categories

• Management
• Report Generation
• Monitoring
• Visualization
Process scheduling server is for Management of Products, Processes and Product
Scheduling.
5.9 Data Dissemination Software (DDS)
It disseminates the generated Internet Products, SIDS products and LRIT/HRIT

products to designated systems.
Following Data Dissemination modes are planned as part of data dissemination
activity
‫ ־‬Transfer of Imager to the web site.
‫ ־‬Conversion of products to GTS format and transmission to Meteorological
communications computer.
‫ ־‬Encoding of products and AWS data in WMO format and transmission to
Meteorological Communications Computers (MCC).
24
6. Geo-physical Product Generation
Parameter Retrieval (PR) and Meteorological maps from INSAT Data Geo-Physical
parameters are retrieved with INSAT-3D/INSAT-3A/Kalpana-1 Meteorological
data on an operational basis in Near Real time and Meteorological Image data
Products generated and Geo-Physical Maps disseminated automatically.
Geo-physical products getsgenerated as soon as Level-1 processing is over. All

geophysical parameters take Level-1 data as input along with dynamic forecast
data, climatological data, DEM and few static inputs.
7. Geophysical (GP) parameter retrieval

Different types of Geophysical parameters Products are retrieved using Imager
and Sounder.
7.1 Imager GP Products : Atmospheric Motion Vector Winds, Outgoing Long

wave Radiation (OLR), Upper Troposphere Humidity (UTH), Sea Surface
Temperature (SST), Quantitative Precipitation Estimates (QPE), Fire, Smoke, Fog,
snow cover and Aerosols}.
The lists of Imager derived geo-physical parameters are given in Table below:
Geophysical parameters from Imager
No. Parameter Input Channels/Data

1 Cloud Mask (CM) MIR, TIR-1, TIR-2
2 Outgoing Longwave Radiation (OLR) WV, TIR-1, TIR -2
3 Quantitative Precipitation Estimation TIR-1, TIR- 2
(QPE) GPI, IMRSA and HE
4 Sea Surface Temperature (SST) SWIR, MIR, TIR – 1, TIR –
2
5 Snow Cover VIS, SWIR, TIR – 1, TIR –
2
6 Fire MIR , TIR -1
7 Smoke VIS, MIR, TIR –1, TIR –2
8 Aerosol VIS, TIR –1, TIR -2
9 Cloud Motion Wind Vector (CMV) VIS, TIR-1, TIR –2
10 Water Vapor Wind Vector (WVWV) WV, TIR-1,TIR –2
11 Upper Tropospheric Humidity (UTH) WV, TIR-1, TIR –2
12 Fog SWIR, MIR, TIR-1, TIR-2
25
7.2 Sounder GP Products :Algorithms and products (e.g. temperature, humidity
profiles, and total ozone) are tested INSAT-3D sounder radiances and operational
products are generated.
Geophysical parameters from Sounder
No. Parameter Input Data

Temperature, Brightness temperatures for 18 Sounder
1 Humidity profile Channel
and Ozone and grey count for channel 19
Sounder retrieved temperature and
2 Geo-potential Height humidity
profiles at 40 pressure levels
Layer Perceptible Retrieved humidity at standard
3 Water pressure levels
Retrieved humidity at standard pressure
4 Total Perceptible Water levels
5 Lifted Index humidity
profiles at standard pressure levels
6 Dry Microburst Index humidity
profiles at standard pressure levels
Maximum Vertical Sounder retrieved temperature and
7 Theta- humidity
E Differential profiles at standard pressure levels
Geo- potential Height and retrieved
8 Wind Index temperature
and humidity profiles at standard pressure
levels
26
8. INSAT-3D Data Products Type and Format
8.1 Product File Naming Convention
All the INSAT-3D products are provided in HDF5 format.
The hdf5 product file name is coined as:
SSNNN_DDMMMYYYY_HHmm_LOP_XXX.h5
Where
SS=Satellite ID (e.g. 3D for INSAT-3D)

NNN=Sensor ID (IMG for Imager, SND for Sounder)
DDMMMYYYY=Date of Acquisition (DD=Day of Month, MMM=Month of the year,
YYYY= year of Pass e.g. 01JAN2014)
HHmm=Time of Acquisition (HH=Hour of day mm=minute of the hour)
LOP=Level of Processing (L1B, L1C, L2B, L2G, L2P, L3B, L3G)
XXX=Parameter Name) or STD or Sector Name
e.g. 3DIMG_01JAN2014_0730_L1B_STD.h5
8.2 INSAT-3D Data Products Types
The various types of data generated by the Data Products System in different
formats are:
‫־‬ LEVEL - 0 (Raw) – for internal use and archival

‫־‬ LEVEL - 1 (Full Globe, Sector)
‫־‬ LEVEL - 2 (Geo-physical)
‫־‬ LEVEL - 3 (Binned Geo-Physical)
The information on HDF, its design philosophy, and its logical and physical
formats, the reader is referred to NCSA web http://hdf.ncsa.uiuc.edu/. The
HDF5 library and utilities (provided by NCSA) can be used to read the contents of
the HDF products. Google KML C++ libraries available at http://code.google.com
are used for writing KML products.
Following section gives INSAT-3D Data Products Types.
27
INSAT-3D Imager Products
S.No. Data Product Processing Code Format Remarks

Level
Standard Products
1 Standard Product Full Disk L1B STD HDF Per Pixel Lat& Lon
as viewed by
Satellite
2 Standard Product Full Disk L1C STD HDF Projected on Fixed

Fixed Grid Grid
3 Standard Sector Product L1C Sector HDF Map Projected

mnemonic
Geo-Physical Parameters
1 Outgoing long wave L2B OLR HDF Per Pixel

radiations
2 Rainfall using Hydro L2B HEM HDF Per Pixel

Estimator
3 FOG L2C FOG HDF Per Pixel
4 SNOW L2C SNW HDF Per Pixel
5 Cloud Mask L2B CMK HDF Per Pixel
6 Upper Troposphere L2B UTH HDF PerPixel

Humidity
7 Sea Surface Temperature L2B SST HDF PerPixel
Geo-Physical Parameters (Point)
1 FIRE L2P FIR KML Point
2 SMOKE L2P SMK KML Point
3 Atmospheric Motion L2P AMV HDF VIS, TIR, WV, MIR
28
Vectors (Point)
Geo-Physical Parameters (Gridded)
1 INSAT Multi-Spectral L2G IMR HDF 0.1 deg x 0.1deg

Rainfall Algorithm (IMSRA)
2 Quantitative Precipitation L2G QPE HDF 1 deg x 1 deg

Estimation
3 Aerosol Optical Depth L2G AOD HDF 0.1 deg x 0.1 deg
Binned Geo-Physical Parameters (Temporally Binned)
1 Outgoing long wave L3B OLR HDF Daily, Weekly,

radiations Monthly and Yearly
Per Pixel
2 Rainfall using Hydro L3B HEM HDF Daily, Weekly,

Estimator Monthly and Yearly
(Per Pixel)
3 Sea Surface Temperature L3G SST HDF Daily, Weekly,

Monthly and Yearly
0.5 deg X 0.5 deg
4 Upper Troposphere L3G UTH HDF Daily, Weekly,

Humidity Monthly and Yearly
0.1 deg x 0.1 deg
5 INSAT Multi-Spectral L3G IMR HDF Daily, Weekly,

Rainfall Algorithm (IMSRA) Monthly and Yearly
0.1 deg x 0.1deg
6 Quantitative Precipitation L3G QPI HDF Daily, Weekly,

Index Monthly and Yearly
(1 deg x 1 deg)
29
INSAT-3D Sounder Products
S.No. Data Product Processing Code Format Remarks

Level
Standard Products
1 Standard Product L1B STD HDF Per Pixel Lat& Lon

as viewed by
Satellite
Geo-Physical Parameters
1 Vertical Profiles and L2B PFL HDF Profile on 3x3

Derived products Pixels (Average)
30
9.IMDPS Storage Infrastructure
The processed data volume of all scans of all channel of INSAT-3A (VHRR,
CCD), KALPANA-1 (VHRR) and INSAT-3D (Imager & sounder) is works out to
be approx. 4 GB per hr. during imaging duration. Assuming 20 hrs.imaging
duration per day the data volume for one month would be around 2.4 Tb. To
cater to this data storage requirement an 11 TB SAN online and 11 TB
Archival SAN along with NAS gateway, 11TB disk library and 48 slot LTO
based tape library (DLT 400/800 GB media) are used in IMDPS storage
system. The storage infrastructure consists of various parts as shown below :
All the components of the storage infrastructure are now explained below:
9.1 Storage Area Network/Online SAN Storage(used in IMDPS system)

In IMD, an EMC CX3-80 model is being used as a SAN server which is used to
connect all the storage devices with the main data processing servers. This
31
SAN server is made the Online FC storage which stores all the live data from
the 3 satellite (i.e. KALPANA-1, INSAT-3A, INSAT-3D).
An EMC CX3-80 model is being used here which is one of the most powerful
midrange storage arrays, it provides high expandability and a higher
application performance. The EMC CX3-80 shares the same hardware and
software architecture as the CX3 model 40 & model 20 but with faster
processors, high capacity, greater connectivity and greater throughput.
The image cannot be display ed. Your computer may not hav e enough memory to open the image, or the image may hav e been corrupted. Restart y our computer, and then open the file again. If the red x
still appears, y ou may hav e to delete the image and then insert it again.
Data Processing System and Storage System Architecture

32
10. Network Clients System
To cater the requirement of user 25 image analysis client workstations, one data
dissemination server, one product monitoring and management system and one
AWPGS server are used as a network client and continuously interact with data
processing system and data storage system.
11. Satellite Image Display System (SIDS)

The Satellite Image Display System (SIDS) is one of the sub systems of INSAT-3D
METEOROLOGICAL Data Processing System (IMDPS). It consists of two HP
XW6600 workstation, TV Display 26”/42”,24 port gigabyte Ethernet LAN switch,
Ku BAND DTH (Including Antenna LNBF & Receiver with remote control) ,Client
Set-top box.The prime objective is to receive the images from Data Dissemination
Service system (DDS) and transmit to TV receivers available at 25 different
locations within the IMD campus. Transmission of signal from SIDS (head-end)
system to LCD TV receiver (user end) using LAN cable network. SIDS system
interacts with DDS to receive the images for transmission.
The Satellite Imagery Display System (SIDS) as the name suggests, is an image
display system capable of:-
 Receiving images from 19 different folders of DDS and transmitting these to

TV receivers (LCD).
 Transmitting images (of different satellite and sensors) to 19 channels and
1DD news channel.
 Display of animated images as a video for showing the movement of clouds.
 Functioning automatically and work (fetching data from DDS and sending to
TV receiver sets at user end) without any user intervention.
SIDS have adding features of scheduling the image delivery system in terms of
setting time interval for searching and sending/ replacing by latest image at
server end on start and stop, crop, adding annotation and scrolling text.
33
Satellite Imagery Display System (SIDS)
34
12. INSAT-3D IMAGER SAMPLE DATA PRODUCTS
INSAT-3D Images and products are generated in real-time and disseminated through
IMD Web to user community.
35
12.1 Imager FULL GLOBE Images
VISIBLE Band
36
SWIR Band
37
TIR1 Band
38
WV Band
39
12.2 Imager Sector Images
ASIA SECTOR RGB NORTH-WEST SECTOR(TIR2)
NORTH-WEST SECTOR (VIS) with District boundary NORTH-EAST SECTOR (MIR) with District Boundary
40
13. INSAT-3D IMAGER GEO-PHYSICAL PARAMETERS
Following Imager geophysical parameters are available in HDF5 format.
 Outgoing Longwave Radiation (OLR)

 Quantitative Precipitation Estimation (QPE)
 Sea Surface Temperature (SST)
 Cloud Motion Vector (CMV)
 Water Vapour wind (WVW)
 Upper Tropospheric Humidity (UTH)
 Snow cover
 Snow depth *
 Fire
 Smoke
 Aerosol
 Fog
 Cloud Mask
41
13.1 Outgoing Long wave Radiation (OLR)
13.1.1 Introduction
The outgoing long wave radiation is a crucial parameter for studying many areas
in the field of atmospheric sciences. The OLR has been used traditionally for
radiation budget studies of the Earth atmospheric system. The OLR also has
been used for the atmospheric circulation studies over tropical region. This is
42
mainly due to the fact that in the tropics, the OLR is largely modulated by
cloudiness. In particular it varies with the cloud top temperature, and
consequently, low values of OLR indicate major convective system.
13.1.2Brief Methodology
Total outgoing long wave radiation (OLR) flux, thermally emitted from earth
atmosphere system, is estimated by applying regression equation relating OLR
flux with Geostationary Indian National Satellite (INSAT) VHRR observed WV (5.7
to 7.1μm) and infrared window radiances (10.5 to 12.5 μm). The coefficients of
the regression equations is determined from results of the radiative transfer
model simulation with various atmospheric conditions. To develop the empirical
relationship between narrow band radiances and OLR a large database of
spectral radiance fields L (θ, λ) was built using the Santa Barbara DISORT
Atmospheric Radiative Transfer (SBDART) model. The radiative transfer
computations were performed at 19400 wavelengths covering the thermal region
3–100 μm and for 4704 realistic conditions of the Earth–atmosphere system.
Genetic Algorithm is used to find the optimum regression equation relating the
broadband flux (OLR) with that of WV (5.7 to 7.1μm) and infrared window
radiances (10.5 to 12.5 μm).
13.1.3 Output Specificatiion
Parameter Unit Min Max Accuracy Resolution Frequency
OLR Wm-2 90 350 2% 1 x 1 degHalf Hourly,Hourly,

Daily,Monthly and
Seasonal.
13.1.4 Format of the output and the domain
Lat, Lon, OLR,
Domain: 40 S to 40 N, 30 E to 120 E
13.1.5 Applications
OLR is an important value for the earth radiation budget. Absorption of solar
radiation and emission of terrestrial radiation drive the general circulation of the
atmosphere and are largely responsible for the earth's weather and climate. It is
also used as one of the parameters for declaring on-set of monsoon over Kerala.
43
13.2. Satellite Rainfall Estimation
There are two main objectives under the rainfall retrieval algorithms from INAST-
3D. These techniques are popularly known as GOES Precipiattion Index (GPI)
and Insat Multispectral Rainfall Algorithm (IMSRA) methods.
13.2.1 Operational Implementation Rainfall Estimation using GPI
Step 1 : Conversion from gray count to BT

Step 2: Histogram Generation:
44
Step 3: Grid-wise Statistics
Step 4: Grid-wise Rain Clod Detection:
Step: 5 Sub-Grid Scale Rainfall
13.2.2Operational Implementation of Rainfall Estimation using IMSRA
Step 1 :Conversion from gray count to BT
Step 2: Cloud Classification using an IR and WV channels.

Step 3: Grid Average of IR TBs (0.250x0.250) to match with microwave rainfall
from TRMM (i.e. INSAT-3D-IR-TBs and TRMM Precipitation Radar rainfall).
Step 4: Calibration of IR brightness temperatures with the contemporary
microwave radiometric measurements of rainfall (e.g. TRMM PR Rainfall ).
Step 5. The model based forecast of precipitable water and relative humidity for
correction of rainfall equation based on above calibration procedure.
Step 6: Validation using DWR and AWS
Step 7. Processing of the data for the different temporal scales
For both GPI and MSRA the domain for rainfall estimation mainly would be the
same but the QPE from GPI and IMSRA are available in 1x1 and 0.25x0.25
degree grids respectively.
ASCII table: latitude , longitude , Rainfall
Domain : 40 S to 40 N, 30 E to 120 E,
Frequancy: Half Hourly,Hourly, Daily,Monthly and Seasonal.
13.2.4 Method of validation
The state of art method for validation of QPE with the ground truth on different
temporal and spatial scale will be carried out mainly with the data of DWR, AWS
and Rain Gauges.
45
13.2.5 Applications
The IMSRA approach is used to optimize the identification of raining clouds

located at a given altitude estimated from the cloud-top temperature. This
product can be used to identify the location of rain/heavy rain quantitatively. It
can also distinguish thick cirrus clouds from deep convective cores.
46
13.3 Operational Implementation of High-Resolution Rainfall Measurements
(Hydro-Estimator)
1. Provides pixel-scale, half-hourly precipitation measurements over land and

oceans.
2. The rain at a pixel is considered to be effected by the brightness
temperature of the surrounding pixels. This helps in determining the
location of the pixel (for example, over active core area or over anvil).
47
3. At a pixel under consideration, the IR measurements are blended with NWP
model fields for more precise rain detection and measurement
4. Dynamically varying brightness temperature – rain rate relationships are
established for convective and stratiform rain at each pixel. These
relationships are determined by pixel location in the storm and available
TPW from NWP model. Rain at a pixel is a combination of both convective
and stratiform.
5. NWP model derived TPW and RH used for rain detection and wet
tropospheric correction (e.g., dryer atmosphere below cloud base results
evaporation of rain drops)
6. NWP model derived 850 mb wind fields laced with surface tropography is
used for enhancement/reduction in rain amount due to orography.
7. Equilibrium level computed using NWP model derived T and RH (or T and
Td) profiles is used to make corrections for rain from warm clouds.
13.3.1Application
The HE uses infrared (IR) window channel (10.7-μm) brightness temperatures as

the main basis for discriminating raining from non-raining areas and for
estimating rainfall rates, and also uses data from numerical weather prediction
models to provide additional information about moisture availability, subcloud
evaporation of precipitation, impact of the thermodynamic profile on cloud
heights, and orographic enhancement or reduction of rainfall.
48
13.4 Sea Surface Temperature (SST)
13.4.1 INSAT – 3D derived SSTs
Sea surface temperature would be derived from split thermal window channels
(10.2-11.3μm, 11.5 –12.5μm) during daytime and using additional mid IR
49
window channel (3.7 – 4.1μm) during nighttime over cloud free oceanic regions.
The most important part of the SST retrieval from IR observations is the
atmospheric correction. Specially over tropics, this atmospheric correction is
dominated by the high variability in vertical distribution of the intervening
atmospheric water vapor. This correction has beendone through suitable
characterization of tropical atmospheres in radiative transfer model to determine
the brightness temperatures of INSAT-3D channels and then generating the
regression coefficients for SST retrieval.
13.4.2 Operational Implementation
Step-1. Determine radiances using the processed Imager data and convert
radiances to temperature using lookup table.
Step-2. Processing for discrimination between radiances from cloud free sea
surface and those from the cloud tops
Step-3. Correction for atmospheric attenuation is taken care in SST computation
for multichannel SST retrieval, but in case of availability of only one thermal
channel, total water vapour field will be required externally from model output.
Step-4. Computation of SST
During daytime, for cloud free pixels, SST is computed as
SST= A0 + A1* T11+ A2*dT + A3*dT2
Where A0, A1, A2, and A3 are satellite zenith angle dependent coefficients
determined by radiative transfer model.
dT = T11 – T12
T11 and T12 are brightness temperatures for the split-window channels.
During Nighttime, for cloud free pixels, SST is computed as
SST= B0 + B1*T3 + B2 *dT + B3 *dT2
Where B0, B1, B2, and B3 are satellite zenith angle dependent coefficients
determined by radiative transfer model.
dT = T11 – T12
T11 and T12 are brightness temperatures for the split-window channels and T3 is
the brightness temperature for channel 3 of IMAGER.
Single channel SST is computed as
SST = a + b*Tb +c*WV
50
Where
Tb is brightness temperature of the IR imager channel, WV is total water vapour

content, a, b and c are regression coefficients generated through radiative
transfer model
Step-5.Quality control/Editing of derived SSTs.
Satellite data are included into the final file if the absolute difference between the
satellite SST and the climatology is strictly below 3 C. The channel 5 standard
deviation must also be lower or equal to 1 C.
Parameter Unit Min Max Theoretical Resolution
Accuracy
SST(day) Kelvin 285 310 1–2K 0.5 x 0.5 deg
SST(night) Kelvin 285 310 <1K 0.5 x 0.5 deg
ASCII table: latitude , longitude , sst
Domain :
40 S to 40 N, 30 E to 120 E ,
Frequancy: Half Hourly, Hourly, Daily, Monthly and Seasonal.
13.4.5 Applications
Sea surface temperature affects the behavior of the Earth's atmosphere above, so
their initialization into atmospheric models is important. While sea surface
temperature is important for tropical cyclogenesis, it is also important in
determining the formation of sea fog and sea breezes.
51
13.5Cloud Motion Vectors (CMV)
13.5.1 Introduction
Wind is a key parameter for weather forecasting, meteorological studies

and climate related applications. As a result of persistent efforts by the scientists
and operational community world over, sufficiently reliable techniques have been
developed for using the imagery data from geostationary meteorological satellites
in order to derive Atmospheric Motion Vectors (AMV) by tracking movement of
clouds and moisture features in successive half hourly images. The globally
derived Atmospheric motion vector (AMV) fields are now well established and
constitute an essential product especially for numerical weather prediction (NWP)
related applications. These data are now being routinely assimilated in
operational models by many NWP centers in the world.
Early attempts to quantify the accuracy level of these winds based on

manual methods to track moisture features in animated image sequences met
with modest success. The water vapor images from geostationary satellites have
52
allowed the estimation of both upper-level moisture flow and the flow that
corresponds to water vapor layers. The resulting product or water vapor wind
(WVWs) has generally an improved coverage with respect to the CMV product, the
major advantage being the availability of many more targets in the image.
There are five types of wind products are being generated at IMDPS from INAST-
3D imager such as Visible(Day)/MIR wind (Night), WVW from WV channel and
CMV from IR1 channel and two blended products of low level wind (using
VIS/MIR channel and WV) and high level winds (WV and IR1 channel).
13.5.2 Operational Implementation CMV at IMDPS
Step 1: Conversion from gray count to BT
In this step a radiation model will be used to convert the instrument measured
radiances into brightness temperature. This requires an accurate definition of the
spectral response of the satellite.
Step 2: Tracer selection from image
Cloud tracer selection will be done by evaluating the maximum local gradients
surrounding each pixel in the target array and selecting the maximum brightness
temperature of the window.
Step 3: Model wind input for tracking
In this step numerical model forecast upper-level winds is required to economize

the search area.
Step 4: Tracking
The tracking employs a simple search for the mean absolute difference of the
radiance difference between the target and search arrays in subsequent half hourly
images. This search will be done in the direction of ± 30º of model wind. The cross-
correlation algorithm will also be employed for search operation.
Step 5: Wind generation
Calculate the wind speed and wind direction. Repeat the Step 1 to Step 4 for 2nd
and 3rd half-hourly images. Again calculates wind speed and direction. If the two
vectors survived the consistency checks, an average will be calculated and this will
become the representative of CMV.
53
Step 6: Initial height assignment
The brightness temperature in the target window will be averaged and matched
with collocated numerical model temperature profile. The level of optimum fit will
be assigned as initial height.
Step 7: 3D objective analysis
3D-objective analysis of the CMVs will be done using background information from
model forecast. Recursive filter (RF) analysis will be employed for object
Step 8: Speed bias adjustment and variational method
At this point an adjustment to CMV speed will be applied based on well known
slow bias in upper-tropospheric winds. This slow bias will be mitigated by an
empirically determined correction term.The CMV heights will be then adjusted by
conducting a vertical search aimed at minimizing a simple variational penalty
function using the initial wind analyses and the model forecast.
Step 9: Final 3D objective analysis
The RF analysis will be repeated by using reassigned CMV in step 8 and output of
the first analysis (step 7) as a background field. The output after this analysis will
provide quality estimates for each vector based the local quality of the analysis.
Vectors which will fail threshold quality will be rejected.
The following parameters will be provided as output of CMV
a) Zonal and meridional components of the wind vectors.
b) Latitudinal and longitudinal position.
c) Height of CMV
The HDF product will be available in the following format:
Parameter Latitude Longitude Level U V Quality Wind Wind
Comp CompFlag Speed Dir
CMVDeg N Deg E hPa m/s m/s % m/s Deg
Frequency: Hourly, 3-Hourly.
54
13.5.4 Method of validation
The evaluation of CMV should be taken into both qualitative and quantitative
measures. Quantitative assessment of the CMV product is possible from
statistical analyses and impact on NWP. The traditional method of validation is
matching observations with collocated radiosondes. The statistical validation will
be done according to the CGMS winds evaluation reporting guidelines. These
statistics can provide a fixed measure of product quality over time and can be
employed in determining observation weight in objective data assimilation. At the
CGMS XXIII the Working Group on Satellite Tracked Winds recommended that
evaluation of operational wind production quality should be accomplished with a
new standardized reporting method. The recommended three parts to the report
are;
i) Monthly means of speed bias and rms vector difference between radiosondes
and satellite winds for low (>700 hPa), medium (700-400 hPa), and high (< 400
hPa) levels together with the radiosonde mean wind speed. This should be done
for three latitude bands : north of 20 N, the tropical belt (20 N to 20 S), and south
of 20 S.
ii) Trends of the evaluation statistics for the monthly cloud motion vectors and
water vapor motion vectors through the last 12 months.
iii) Information on recent significant changes in the wind retrieval algorithm.
13.5.5 Applications
The main application of the satellite-derived cloud-motion vectors is their use as

winds in the data analysis for numerical weather prediction. At data sparse
region they constitute an indispensable data source for numerical weather
prediction. The CMV`salso gives the indication of trough, ridges, circulation,
steering current jetstream etc.
55
13.6Water Vapor Wind Vector (WVWV)
13.6.1 Introduction
In early nineties, forecasters have used extensively the water vapor radiances
from geostationary satellites qualitatively. In quantitative studies, the first
attempt was made by manually tracking moisture features in automated image
sequences. Later studies on this area have shown much maturity with ability of
automated wind extraction methods for producing spatially coherent Water Vapor
56
Wind Vector (WVWV) fields comparable in quality to operational cloud tracked
winds. The U.S Geo-stationary Operational Environmental Satellites (GOES-8 etc
series) are equipped with one imager and two sounder water vapor channels. As
of GOES-13 the spatial resolution of the WV imager has been improved from 8Km
to 4 Km at the sub-satellite point, and the radiometer is spectrally wider with a
central wavelength of 6.5 micro m instead 6.7 micro m. The sounders look
progressively deeper into the troposphere as the spectral band wavelength moves
away from the 6.3micro m absorption band center. This facts opens the
opportunity for a 3-dimensional reconstruction of atmospheric motions. Water
vapor winds from image data taken by the WV channel aboard the Japanese
satellite Geo stationary Meteorological Satellite (GMS)-5 have been produced
since 1995, and clear-sky WV segments have been separated since 1998.
Step 1 : Conversion from gray count to BT
In this step a radiation model will be used to convert the instrument measured
radiances into brightness temperature. This requires an accurate definition of the
spectral response of the satellite.It is important that the different instruments are
sensing radiation from slightly different parts of H2O absorption band. Sampling
the centre of the absorption band yields radiation from the upper levels of the
troposphere, similarly sampling away from the centre of the absorption band
yields radiation from lower levels of the atmosphere.
Step 2: Tracer selection from image
Water-vapor images tracer selection will be done by evaluating the bidirectional

gradients surrounding each pixel in the target array and selecting the maximum
brightness temperature of the window exceeding some thresholds value.
Step 3: Model wind input for tracking
In this step numerical model forecast upper-level winds is required to economize

the search area.
Step 4: Tracking
The tracking employs a simple search for the minimum in the sum of squares of
the radiance difference between the target and search arrays in subsequent half
hourly images. This search will be done in the direction of ± 30º of model wind.
The cross-correlation algorithm will also be employed for search operation.
57
Step 5: Wind generation
Calculate the wind speed and wind direction.Repeat the Step 1 to Step 4 for 2nd
and 3rd half-hourly images.Again calculate wind speed and direction. If the two
vectors survived the consistency checks, an average will be calculated and this
will become the representative of WVWV.
Step 6: Initial height assignment
The brightness temperature in the target window will be averaged and matched
with collocated numerical model temperature profile. The level of optimum fit will
be assigned as initial height.
Step 7: 3D objective analysis
3D-objective analysis of the WVWV will be done using background information

from model forecast. Recursive filter (RF) analysis will be employed for objective
analysis.
Step 8: Speed bias adjustment and variational method
At this point an adjustment to WVWV speed will be applied based on wellknown
slow bias in upper-tropospheric winds. This slow bias will be mitigated by an

empirically determined correction term. The WVWV heights will be then adjusted
by conducting a vertical search aimed at minimizing a simple variational penalty
function using the initial wind analyses and the model forecast.
Step 9: Final 3D objective analysis
The RF analysis will be repeated by using reassigned WVWV in step 8 and output
of the first analysis (step 7) as a background field. The output after this analysis
will provide quality estimates for each vector based on the local quality of the
analysis. Vectors which will fail the empirically determined threshold quality will
be rejected.
Format of the output and the domain

As output of WVWV the following parameters will be provided to IMD:
a) Zonal and meridional components of the wind vectors.
b) Latitudinal and longitudinal position.
c) Height of WVWV
The format of the final product will be like this:
58
Parameter Latitude Longitude Level U V Quality Wind Wind
Comp CompFlag Speed Dir
WVWVDeg N Deg E hPa m/s m/s Flagm/s Deg
Frequency: Hourly, 3-Hourly.
13.6.3 Methods of validation
Similar to CMV and other winds.
13.6.4 Applications
It has been demonstrated by studies that, the upper level winds derived from
satellites have proved to be very useful for predicting the future track position of
depressions and well marked low pressure areas with deep vertical extent. On the
basis of their potential use for better future track predictions it is possible to give
more accurate heavy rainfall warnings to the areas likely to be affected by these
weather systems. It is possible to give more precise warnings to the affected areas
at least 24 to 48 hours in advance since these types of weather systems are
steered by the upper level winds.
59
13. 7Upper Tropospheric Humidity (UTH)
13.7.1 Introduction
Upper Tropospheric Humidity (UTH) is an estimate of the mean relative humidity

of the atmosphere between approximately 600 hPa and 300 hPa. UTH is basically
a measure of weighted mean of relative humidity according to the weighting
function of the water vapour channel. Therefore, UTH is more likely a
representative of the relative humidity around the atmospheric layer where
weighting function of water vapour channel peaks.
13.7.2 Brief Methodology
UTH formulation is based on the analytical expression developed by Soden and

Bretherton [1996],
ln[UTH . po/ cos(θ)] = a + b . Tbwv
60
where, po is the normalized or scaled reference pressure and is equal to the
pressure of the level at temperature 240K divided by 300 hPa, i.e., po=
p(T=240K)/300. Standard tropical atmospheric profile of temperature is used, as
INSAT-3D a geostationary satellite making observations over most of the tropical
region extending up to the mid-latitudes. A training dataset of atmospheric
profiles is generated using standard tropical temperature profile and different
values of relative humidity above 800 hPa ranging from 1% to 100% with the
interval of 1% at the lower values and 5% at higher values. Satellite zenith angle
was varied from θ = 0° to θ = 65°, such that the interval sec(θ) = 0.1 to keep
equal interval in the path lengths. Empirical coefficients are computed from this
simulated dataset.
Cloudy pixels are identified and excluded from further calculations using
standard cloud detection algorithm. In the 5 x 5 pixel box, if the number of
cloudy pixels is more than 25 % than the brightness temperature of clear pixels
are averaged. Using the zenith angle, p0 (computed empirically using a
polynomial function of latitude and month) and the UTH coefficients for
corresponding satellite and month of the observation, UTH is calculated using the
clear WV channel brightness temperature.
13.7.3 Output
Parameter Unit Min Max Accuracy Resolution
UTH % 0 100 30 % 0.5° X 0.5°
Frequency: Half Hourly, Hourly, Daily and Monthly
Binary file giving details of the time and location of the retrieval, zenith angle, %
of clear pixels used in averaging within 5 x 5 pixel box, and UTH (%).
Domain: 40 S to 40 N, 30 E to 120 E
13.7.5 Applications
Upper tropospheric humidity (UTH) is a crucial parameter for meteorology and

climate research. The UTH are used together with global cloud imagery to study
the relationship between deep convection and moisture in the upper troposphere.
61
13.8Snow-Cover Mapping
13.8.1 Introduction
The snow-mapping algorithm will use a grouped-criteria technique using the

Normalized Difference Snow Index (NDSI) and other spectral threshold tests to
identify snow on a pixel-by-pixel basis, and to map snow cover in dense forests.
The NDSI is usefulness for snow mapping, as it reflects more in the visible than
in the short-wave IR part of the spectrum. In addition, the reflectance of most
clouds remains high in the short-wave IR, while the reflectance of snow is low.
Validation of the INSAT snow maps will be carried out using snow maps
generated using AWiFS data of Resourcesat and a limited amount of field
measurements. In addition, validation will also be carried out using visual
62
interpretation and the MODIS snow maps. The accuracy of the snow maps will
vary with land-cover type. Hence, the Snowmap algorithm has been and will
continue to be tested for a variety of land covers. Error estimates have been
determined from field measurements for different land covers, and these errors
are used to estimate the expected maximum monthly and annual errors in
Himalayan snow mapping using the algorithm.
Step 1 : Conversion from gray count to Reflectance and BT
In this step, the DN values will be converted into Radiance, Reflectance and
Brightness temperature using calibration coefficient available with INSAT 3D
data. The radiometrically processed data will also be geolocated and this will
provide a radiometrically and geometrically corrected product to be used further
in algorithm.
Step 2 : Cloud elimination
This step includes the cloud screening from the INSAT 3D image. Cloud
screening is based on the temperature data and will help to clear cloudy pixels
which possibly are not feasible to screen out using visible and SWIR channel
based NDSI image.
Step 3 : Generation of NDSI image
Normalized Difference Snow Index will be computed using the normalized ratio of
visible and SWIR channel as given below;
NDSI = Rfl (visible) – Rfl (SWIR) / (Rfl (visible) + Rfl (SWIR)
Step 4: Ortho rectification
It will provide area on slope and provide correct area of snow cover. This will
generate a daily snow cover map. Composite images of 10 days will be used to
generate a 10 Daily snow cover map and this map will provide maximum snow
areal extent. Snow pixel will be identified in 10 daily products, if occurs on any
single day.
SCM Km2 - - - 1 x 1 deg

63
Geolocated Snow map & Areal extent of snow cover in Himalayan region.
40 Sto 40 N; 30 E to 120 E
Frequency: Half Hourly
13.8.5 Applications
The highly reflective nature of snow combined with its large surface coverage
makes snow an important determinant of the Earth's radiation balance. Many
areas of the world rely on snowmelt for irrigation and drinking water and must
monitor snow packs closely throughout the winter and spring for assessment of
water supply.
64
13.9Snow Depth Estimation
13.9.1 Introduction
Important step in estimation of snow depth is generation of snow fraction map.

Since Himalayan region is highly mountainous, the data processing would
initially need calibration, georeferencing, atmospheric correction, snow pixel
detection, cloud masking, local illumination angle correction, diffuse radiation
correction, BDRF correction and orthorectification. This will be followed by Linear
or non-linear mixing modeling to estimate snow fraction in each pixel. The mixing
modeling would need large spectral reflectance library of snow and other land
features of Himalayan region, to identify end members of snow and snow free
terrain. Next important step in the investigation is to develop a relationship
between snow fraction and depth. This would need field snow depth data. In
North America this relationship was developed using 1400 field observatories.
Relationship was found useful; if snow depth is less than 27 cm. Field data needs
to be collected from India Meteorological Department.
Step 1: Conversion from gray count to BT
In this step, the DN values will be converted into Radiance, Reflectance and
Brightness temperature using calibration coefficient available with INSAT 3D
data. The radiometrically-processed data will also be geolocated and this will
provide a radiometrically and geometrically corrected product for further
analysis.
Step 2: Cloud elimination
This step includes the cloud screening from the INSAT 3D image. Cloud
screening is based on the temperature data and will help to clear cloudy pixels,
which possibly are not feasible to screen out using visible and SWIR channel
based NDSI image.
Step 3: Topographic and Atmospheric correction
In this step, DEM data will be used to compute the terrain parameters like slope
and aspect of the terrain, which will further be used to compute the local
illumination angle. Atmospheric correction algorithm will be applied to compute
the atmospherically corrected radiance and reflectance values with standard
procedures. This step will provide us the atmospherically and topographically
corrected radiances and reflectance.
65
Step 4: Snow pixel identification
Snow pixel identification will be done by using NDSI based approach which uses
thenormalized ratio of visible and SWIR wavelength reflectance. This prepares a
binary mapof either snow or non-snow pixels.
Step 5: Snow fraction retrieval
Fractional snow map will be prepared by using linear mixing approach. This
considers land and pure snow as end members and computes the snow fraction
of any particular pixel using reflectance value.
Step 6: Snow Depth Estimation
A statistical relationship was developed between snow fraction and depth

(Romanov and Tarpley, 2004). Snow depth will be estimated by using this
relationship for each pixel.
Parameter Unit Min Max Resolution
SD cm - - 1 x 1 deg
Geolocated Snow depth map.
40 S to 40 N; 30 E to 120 E.
13.9.5 Applications
Three of the most important properties of a snow cover are depth, density, and
water equivalent. If the snow depth and density are known, then the snow-water-
equivalent (SWE) may be calculated. SWE is a hydrologically important
parameter as it determines the amount of water that will be available as
snowmelt.
66
13.10 FIRE IDENTIFICATION
13.10.1 Introduction
Forest fire management in tropical countries is one of the major issues. Satellite
data with suitable spectral bands, good temporal resolution and bare minimum
spatial resolution, which can locate fire, can play an important role in
development of country’s forest fire management system. One of the most
important critical elements of the forest fire management system in the country is
the real time detection of fire and its progression monitoring; study the rate,
direction and quantitative estimation of fire spread. Geostationary satellite like
67
INSAT-3D with imager data of 4x4 km. in MIR and in TIR-I and TIR-II will help in
detecting and monitoring of large scale forest fires in Indian subcontinent.
The fire detection algorithm employs multi channelthresholding contextual

technique. The algorithms will first masks out ocean, desert, water and cloud
pixels and then proceed further. The contextual algorithms identify a fire pixel
based on the level of contrast between the potential fire pixel and its
“background” pixels (the definition of background varies according to kernel size).
An initial threshold test will reduce the number of candidate pixels to be applied
to the contextual algorithm surrounding the subject pixel values. The initial
thresholds are more liberal to avoid missing real fires. The second test computes
the mean and standard deviation of the threshold variables from non-potential
fire pixels surrounding a potential fire pixel. After obtaining these statistics, they
are used to re-define the thresholds to confirm a fire. Background temperature
will be updated with a separate algorithm which will use TIR-1 and TIR-2
Channels of INSAT-3D. The fine tuning of threshold values and kernel size will be
done after validation experiments for known fire pixels.
Output: lat ,lon, Fire flag (Image)
Fire flag: 1 - Fire pixels, 0 - Non-fire pixels
Resolution: 4 x 4 Km
Lat, Lon, Fire,
Domain:
40 S to 40 N, 30 E to 120 E
13.10.5 Applications
To reduce impacts of forest fire in boreal forest area, the early fire detection is one
of essential components in firefighting activity because of difficulties of fire
suppressing in remote area without water.
68
13.11Smoke Identification (SI)
Smoke is a form of particulate matter, which contains liquid or solid particles of

the size ranging from 1−200μm. It is formed by combustion or other chemical
processes. Each year more than 100 million tons of smoke aerosols are released
into the atmosphere as a result of biomass burning. More than 80percent of this
burning is in the tropical regions. Smoke plumes can travel over hundreds or
69
even thousands of kilometers horizontally and also reach up to stratosphere
under certain atmospheric circulation conditions. Thus smoke can have an
impact far beyond the region of fire activity. Smoke plays a major role on the
radiation balance of the earth-atmosphere system. Smoke particles scatter and
absorb incoming solar radiation, thereby having a two-fold impact, i.e. a cooling
effect at the surface, but warming effect on the atmosphere. Since the magnitude
of the scattering effect outweighs that of absorption, smoke has a net cooling
effect at the top of the atmosphere-surface system. This is often called the direct
effect of smoke aerosols. Smoke can also modify the short wave reflective
properties of clouds by acting as cloud condensation nuclei. Under a limited
supply of water vapor, an increased number of nuclei result in smaller cloud
droplets that have higher reflectivity than larger cloud droplets .This effect called
the indirect radiative forcing, is difficult to quantify and has large uncertainties
associated with the sign and magnitudes. Understanding such numerous and
complex effects of smoke on weather and climate requires a good knowledge of
the spatial and temporal variation of smoke and its optical properties, which is
only feasible by means of satellite observation.
13.11.2 Operational retrieval implementation
The operational implementation of the retrieval process can be enumerated as

follows:
1. Conversion of gray counts to radiance and brightness temperatures
2. Conversion of gray counts to Albedo for the VIS channel
This is done with the help of conversion tables.
3. Calculation of solar reflectance angle
To exclude sun glint processing path is taken for solar reflectance angle θr
greater than threshold value (approximately 20deg)],
where θ is the viewing zenith angle, Φ is the azimuth angle and, θ0 is the Solar
Zenith Angle.
4. Defining of thresholds
Using training dataset of forest fires over the Indian region for VIS, MIR , TIR-1
and TIR-2 channels, thresholds have to be defined for tests listed above. The
methodology to be followed is to first assume initial value of thresholds based on
literature and then fine tune them, using visual inspection.
70
5. Implementation of five pass tests
6. Cloud edge detection algorithm
A difference image is produced from 10.7 and 3.9μm channels and an adjacent
pixel test is performed to detect cloud edges.
13.11.3 Output
Generation of smoke map
Output: lat ,lon, Smoke flag
Smoke flag: 1 - Smoke pixels, 0 - Non-smoke pixels
Domain : 40 S to 40 N, 30 E to 120 E
Frequency: Half Hourly, Daily and Monthly
Identification of smoke on satellite imagery is a prerequisite to study and retrieve

physical, chemical, and optical properties of smoke.
71
13.12 AEROSOL OPTICAL DEPTH (AOD)
Aerosols play an important role in numerous aspects of human life. Aerosols have
large-scale effects, such as their impact on climate by redistributing solar
radiation and interacting with clouds. Aerosol information is also critical for
atmospheric correction algorithms for multi-spectral satellite sensors and
military operations. The climate effects of atmospheric aerosols may be
comparable to CO2 greenhouse effects, but with opposite sign and larger
uncertainty. Aerosols have a significant impact on human life beyond the climate
72
element. When in the lower troposphere, aerosols cause poor air quality,
reduction of visibility, and public health hazards. Satellite remote sensing
provides a means to derive aerosol distribution at global and regional scales.
Step 1 : Conversion from gray count to Visible radiance and BT

Step 2 : Cloud and sun glint elimination
Step 3 : Generation of surface reflectance image using darkest observations using
last 20 days data
Step 4: Atmospheric correction of INSAT-3D visible image for molecular scattering
and gaseous absorption
Step 5 : Correction for surface reflectance
Step 6: Use LUTs for the inversion of aerosol optical depth (AOD)
13.12.3 Output
Aerosol Dimensionless 0 2 15-20% 4Km
Optical depth at 650 nm
Frequency: Half Hourly, Daily and Monthly
Lat , Lon, AOD ,
Domain :
40 S to 40 N, 30 E to 120 E
Atmospheric correction of remotely sensed surface features, Monitoring of

sources and sinks of aerosols, Monitoring of volcanic eruptions and forest fire,
Health and Environment, Air quality and Climate Change
73
13.13FOG
13.3.1 Introduction
Fog affects visibility near the surface and hence is an important parameter for
aviation, transport on land and sea. Its detection and monitoring by means of
satellites is an upcoming area of research. General methodology involves
identifying some threshold radiances or brightness temperatures at different
wavelengths which can distinguish fog from other cloud and surface features.
Night time fog detection is done by looking at the 10.8 and 3.9μm channel
brightness temperatures, in analogy to the method applied to GOES data. This
technique relies on fog pixels displaying higher brightness temperature
differences as compared to clear pixels and those covered by other clouds. This
74
technique is very efficient in detecting fog during night time. Identifying fog
during day time is a bit complex. A variety of methods are being tried. But with
the radiance measurements very limited in the visible region, many of these
methods are not applicable for INSAT. In view of these one can try the use of
same infrared channels that are used for night time for day time too. However
difference between the two channel brightness temperatures alone is not
sufficient. The threshold identification is dynamic and depends on the solar
zenith angle, local surface albedo and a host of radiative transfer model
simulations with various fog and cloud properties. In short the crux of the
method is elimination of possibility of different cloud types before coming to the
conclusion that the pixel contains fog. Neither day time nor night time algorithm
works during dusk and dawn.
Night Time Fog Detection
 Conversion of grey count to Brightness temperature

 Difference between 10.8 and 3.9μm channel brightness temperatures.
 Cloud elimination (In the above step if difference is greater than 2.5 they
are assumed to represent opaque clouds).
 Fog detection (Intermediate temperature differences between 0.5 and 2.5
K).
Day Time Fog Detection
 Gross cloud check (dynamical threshold between 10.8 and 3.9μm).

 Ice cloud check (estimated using 3.9μm channel brightness temperature).
 Spatial clustering of resulting water clouds.
 Brightness temperature of these clusters compared with adjacent cloud
free pixels to identify if a cloud cluster is within 2000 meters above ground
to identify low clouds.
 For all low cloud clusters the homogeneity of the cloud top height is
tested.10.8 μm brightness temperature is taken as a proxy for altitude.
Where its variation within a cloud cluster remains below a pre-defined
maximum standarddeviation, the cloud cluster is identified as fog/low
stratus.
75
FOG Flag 0 (not a fog) 1 (fog) - 0.5 x 0.5 deg
Frequency: Hourly
Lat, lon, fog
Domain: 40 S to 40 N, 30 E to 120 E.
13.13.5 Application: It can useful for aviation and transport advisory especially
at airports.
76
13.14 Cloud Mask (CMK)
The discrimination of cloud from the clear sky is necessary for many geophysical
parameter retrieval from INSAT data. Clouds are generally characterized by higher
reflectance and lower temperature than the underlying earth surface. As such,
simple visible and infrared window threshold approaches offer considerable skill in
cloud detection. However, there are many surface conditions when this
characterization of clouds is inappropriate, most notably over snow and ice.
Additionally, some cloud types such as thin cirrus, low stratus at night, and small
cumulus are difficult to detect because of insufficient contrast with the surface
radiance. Cloud edges cause further difficulty since the instrument field of view
will not always be completely cloudy or clear.
INSAT-3D VHRR will measure radiances in one visible and one SWIR band at 1 km
spatial resolution, one MIR and two TIR bands at 4 km resolution, and one WV
band at 8 km resolution. Radiances from 3 IR spectral bands TIR –1 , TIR –2 and
MIR which are of same resolution of 4km will be used in the INSAT cloud mask
algorithm to estimate whether a given view of the earth surface is unobstructed by
clouds.
The basis of the methodology is, first to generate a clear composite of the
maximum brightness temperature in thermal channels to get rough idea about
the surface temperature in the clear sky condition at a particular time. This will
be the reference background temperature on which the threshold for a particular
location will be determined to discriminate cloud. Several test like BT threshold
test, Difference Test ( BT11 – BT3.7), Difference Test ( BT3.7 – BT12), Spatial
variability test , Spatial Uniformity test, Adjacent pixel test, Temporal Uniformity
test, Final threshold test will be applied on each pixel to find our whether the
pixel is cloudy, clear, partially cloudy or partially clear.
Parameter Resolution
CMK pixel
77
Scan , pixel, cloud flag
If cloudflag = 0 , pixel is clear
If cloudflag = 1 , pixel is cloudy
If cloudflag = 2, probably clear
If cloudflag = 3, probably cloudy
If cloudflag = 9, Cold space
Domain :
Image Domain
13.15.5Application: It can be used for detection and removal of clouds from

Imager and Sounder payloads.
78
14. SOUNDER ATMOSPHERIC PARAMETERS
BT Image (SWIR6)
79
Temperature, Moisture profile from INSAT-3D Sounder
INSAT3-D Sounder Profile over the Indian region
80
14.1 Introduction
INSAT-3D will carry an 18-channel infrared Sounder (plus a visible channel)

along with a 6 channel Imager. This algorithm is designed for retrieving vertical
profiles of atmospheric temperature and moisture along with total column ozone
content in the atmosphere from clear sky infrared radiances in different
absorption bands observed through INSAT-3D. INSAT-3D Sounder channels are
similar to those in GOES-12 Sounder and many of the spectral bands are similar
to High resolution Infrared Radiation Sounder (HIRS) onboard NOAA ATOVS.
Hence, present algorithm for INSAT-3D Sounder is adapted from the operational
HIRS and GOES algorithms developed by Cooperative Institute for Meteorological
Satellite Studies (CIMSS), University of Wisconsin.
14.2 Algorithm
The algorithm includes generation of the hybrid first guess atmospheric profiles
using a linear combination of regression retrieval and NWP model forecast. This
is followed by a non-linear physical retrieval procedure (Li et al. 2000; Ma et al.
1999) to make the first guess consistent with the Sounder observations. The
PFAAST atmospheric radiative transfer model (Hanon et al., 1996) has been used
for the forward computation of sounder channel radiances alongwith the
Jacobians, which is used in the physical retrieval. The regression coefficients are
computed for 150 zenith angle classes (0 to 65° with increment of sec(θ) = 0.01)
for 3 latitude zones (20S-20N, 20N/S-40N/S, 40N/S-60N/S) from radiosonde
database (SeeBor dataset, University of Wisconsin) separately for land and ocean.
The set of predictors include sounder channel brightness temperature (Tb) and its
quadratic term (Tb2) along with the surface pressure (Ps) from NWP forecast. The
atmospheric profiles obtained from regression retrieval is combined with the NWP
forecast profiles to generate a hybrid regression profile that is used as first guess
for the physical retrieval. The physical retrieval procedure involves minimization
of the following cost function:
J(X)=[Ym – Y(X)]T .E-1.[Ym – Y(X)] + (X - X0)T .H-1.(X – X0)
Where, X is the atmospheric profile vector, X0is the first guess, Y is the
observation vector, Y(X) is the forward model, E is observation error covariance
matrix, and H is the a priori matrix that constrains the solution, usually the first
guess error covariance matrix.
81
14.3 Output Specification
Following parameters will be retrieved from INSAT-3D Sounder:
Parameter Resolution Expected

Accuracy
Temperature profile 30 km x 30 km (3 x 3 Pixels)

30-vertical pressure levels* 1 - 2 °C
Water vapour profile 30 km x 30 km (3 x 3 Pixels)

21-vertical pressure levels
upto 100 hPa ~30%
Ozone profile 30 km x 30 km (3 x 3 pixels)

40 vertical pressure levels
Total Column Ozone 30 km x 30 km (3 x 3 pixels) ~ 5-10 % Dob unit

Surface skin temperature 30 km x 30 km (3x 3 pixels) ~ 0.5 –1 °C
• Algorithm has option for retrieved at 30 km (3 x 3 pixels) and 10 km

resolution (each pixel).
Vertical Pressure Levels (40) in hPa :
1000, 950, 920, 850, 750, 700, 670, 620, 570, 500, 475, 430, 400, 350, 300, 250,
200, 150, 135, 115, 100, 85,70,60, 50, 30, 25, 20, 15, 10, 7, 5, 4, 3, 2, 1.5, 1, 0.5,
0.2, 0.1
14.4 Format of the output and the domain
Direct access binary format with each record giving details of time of
observations, location, forecast profiles, regression first guess profiles, final
physical retrieval output profiles, surface skin temperature, total column ozone,
and other auxiliary information. These retrievals will be made within 60 degrees
from the sub-satellite point.
82
15. Application parameters from Sounder derived products
 Geopotential height,
 Layer and total precipitable water,
 Lifted index from sounder,
 Dry microburst index,
 Maximum vertical theta-e differential,
 Wind index
83
15.1 GEOPOTENTIAL HEIGHT
84
15.1 Introduction
Geopotential height fields (GPH ) at a pressure level indicate the source and sinks
of air masses. Generally high pressure areas have relatively higher geopotential
heights than low pressure areas. It is also a fundamental quantity that is used in
atmospheric general circulation models. At mid-latitudes they also indicate the
underlying large-scale waves such as Rossby waves. By looking at the
geopotential heights one can infer the first guess atmospheric wind pattern (also
known as geostrophic approximation) especially for mid-latitudes.
15.2 Methodology :
Given surface pressure, and temperature profile, height at any pressure level or
thickness of atmospheric layer from surface to a particular pressure level is
computed using the formula:
Rd Tv P
Z ln( s ) Unit: meters
g P
Ps – Surface Pressure, P – Given pressure where geopotential height is to be

calculated, Tv - Layer mean virtual temperature.
15.3 Output Specificatiion
GPH Meters - - +/- 10 m 0.5 x 0.5 deg
Frequency: Hourly
15.4 Format of the output and the domain
Latitude, Longitude, Pressure (1000 to 0.1 hPa), GPH.

Domain: 40 S to 40 N, 30 E to 120 E.
Vertical Pressure Levels (40) in hPa :
1000, 950, 920, 850, 750, 700, 670, 620, 570, 500, 475, 430, 400, 350, 300, 250,
200, 150, 135, 115, 100, 85,70,60, 50, 30, 25, 20, 15, 10, 7, 5, 4, 3, 2, 1.5, 1, 0.5,
0.2, 0.1
15.5 Application: It can be used to analyze the structure of the tropical

cyclones.
85
15.2 Total and Layer Precipitable Water
15.2.1 Introduction
Water vapor content in the atmosphere modifies the air mass characteristics.
Presence of air mass lightens the air mass thereby increasing the potential for
convective activity. As an air parcel rises, it cools dry adiabatically until
saturation vapor pressure is reached. Once saturation vapor pressure is reached
some of the water vapor condenses forming liquid droplets, and releasing latent
heat. This latent heating reduces the amount of cooling and makes the air parcel
morebuoyant. This information is very essential for accurate prediction of
weather and goes as an input parameter in numerical weather prediction models.
86
15.2.2Methodology
Layer precipitable water may be computed using the formula

p2
q
PW   dp
p1
g
where p1 and p2 are bounding pressures of each layer in Pa, q – specific

humidity in Kg/Kg. Total precipitable water is also computed from the same
formula with p1 as surface pressure and p2 as top of the atmosphere pressure
(i.e. about 100 hPa beyond which water vapor amount is assumed to be in
negligible). Unit of precipitable water is mm depth of equal amount of liquid water
above a surface of one square meter (if pressure is in Pa and specific humidity is
in Kg/Kg).
PW mm 0 - 25% 0.5 x 0.5 deg
Frequency: Hourly
lat, lon, PW (at three layers) and TPW (total precipitable water vapor).
Domain: 40 S to 40 N, 30 E to 120 E.
Perceptible water can also been observed at three different layers. The distinct
layers are
Low level preciptable water vapor (LPW) - 1000 hPa to 900 hPa
Mid levelpreciptable water vapor - 900 hPa to 700 hPa
High level preciptable water vapor - 700 hPa to 300 hPa
87
Low and Mid LevelPreciptable Water Vapor
88
High LevelPreciptable Water Vapor
15.2.5 Applications: Layer & Total precipitable water products can provide
additional details about tropical cyclone structure such as asymmetries and
moisture gradients that aid in interpreting tropical cyclones interaction with
dry/moist air.
89
15.3 LIFTED INDEX
15.3.1 Introduction
Lifted index is an indicator of convective activity. It is calculated in the following

steps.:
1. Mean boundary layer (usually lowest 100 hPa layer) temperature and humidity
are calculated.
2. A parcel of air with the above calculated temperature and humidity is lifted
from the middle of the boundary layer dry adiabatically up to lifting
condensation level and then moist adiabatically up to 500 hPa.
3. The environmental temperature minus the parcel temperature at 500 hPa
calculated in the above step is the lifted index (LI).
90
If the parcel temperature is warmer than the environmental temperature it
indicates that the parcel can rise further on its own and LI will be negative.
Based on a large number of LI calculations from radiosonde observations, it is
observed that LI less than -5 implies a very strong likelihood of thunderstorm
activity.
LI Deg C - - 2-3 degC 0.5 x 0.5 deg
Frequency: Hourly
lat, lon, LI
Domain: 40 S to 40 N, 30 E to 120 E.
15.3.4 Applications
Time sequence of geographical pattern of LI may be monitored to study large

scale convective activity and its relation to synoptic circulation systems.
91
15.4. DRY MICROBURST INDEX
15.4.1 Introduction
Evaporative cooling of falling rain is often a much larger effect than the liquid
water loading. In regions such as the western Great Plains of the United States
(e.g. near Denver), the environmental air is often so dry that evaporative cooling
causes dangerous downdraft called downbursts. Hazardous downbursts can
occur even under cloud bases where precipitation evaporates before reaching the
ground. The smaller-diameter but intense downbursts are called microbursts.
Downbursts of 0.5 to 5 km in diameter have been observed. For extreme cases,
downdraft speeds of nearly 10 m/s have been observed 100 m above ground.
This is particularly hazardous to landing and departing aircraft, because this
92
vertical velocity can sometimes exceed aircraft climb rate. Doppler radars can
detect some of the downbursts and give early warning to pilots (Roland Stull,
2000). In India dry desert regions, elevated plains, mountain regions where air is
very dry may be susceptible to microburst.
15.4.2 Methodology
Dry microburst index is generally calculated using the formula,
DMI = Γ + (T −Td) 700− (T − Td) 500
Γ - lapse rate (0C km-1) of the layer from 700 hPa to melting level (i.e. layer at 0o
C), T – Temperature (0C), Td – Dew point (0C). Usually DMI is not calculated for
any retrieval unless the following three conditions are satisfied:
(1) Γ > 6 K km-1 ( somewhat stable for convective activity)
(2) (T - Td)700>= 8 K (implies a very dry atmosphere close to surface)
(3) (T-Td)500<= 8 K (implies some level of saturation at this level)
Suitability of these conditions needs to be studied for the Indian region.
DMI - - 2-3 degC 0.5 x 0.5 deg
Frequency: Hourly
Lat, Lon, DMI
Domain: 40 S to 40 N, 30 E to 120 E.
15.4.5 Applications
Generally dry microburst occurs in situations characterized by high convective

cloud bases and strong evaporational cooling in the sub-cloud layer, resulting in
little or no precipitation at the surface
93
15.5 MAXIMUM VERTICAL THETA-E DIFFERENTIAL
15.5.1 Introduction
Equivalent potential temperature (θe) is the potential temperature that a
saturated air Parcel would have if raised moist adiabatically to the top of the
atmosphere. It is given by the formula
Lq
 e   exp( )
C pT
Where θ - Potential temperature, L – Latent heat of condensation, q – Specific

humidity, T – Temperature, Cp – Specific heat of dry air at constant pressure. As
the moisture decreases with height, θe also decreases with height and reaches a
minimum in the middle troposphere, then increases again into the upper
troposphere. A quantity called dry microburst potential index (MDPI) is calculated
by dividing thetaed by 30. As this formula for MDPI is empirical, its suitability
over the Indian region needs to be studied.
15.5.2 Methodology
From the retrieved temperature and humidity data calculate theta-e at standard
pressure levels using the equation given in the previous section. Calculate the
difference between the lowest level theta-e and subsequent levels up to 500 hPa.
The maximum of these differences is maximum vertical thetae difference
(thetaed).

THETAED ºC - - 2-3degC 0.5 x 0.5 deg
Frequency: Hourly
Lat, lon, thetaed
Domain: 40 S to 40 N, 30 E to 120 E.
15.5.5 Applications: The maximum vertical θe differential between the boundary

layer and middle troposphere is a measure of atmospheric instability in the
vertical direction. Larger the theta-e differential, more unstable is the atmosphere
in the vertical direction.
94
15.6 WIND INDEX
15.6.1 Introduction
Wind Index (WI) is a parameter based on vertical equations of momentum and

continuity with certain simplifying assumptions. It is given by the formula
WI  5[ H M RQ (T 2  30  QL  2QM ]1 / 2
Where WI – Maximum wind gust (knots, at the surface)
HM – Height above ground of melting level (in km)
RQ – QL/12 but not >1, QL – Mean mixing ratio (g/Kg) in lowest 1 km
T – Lapse rate (oC km-1) from surface to melting level, QM – Mixing ratio at
melting level.
15.6.2 Methodology
1) From the retrieved temperature profile at standard pressure level, the pressure
95
level at which temperature reaches 0o C is estimated by means of interpolation.
Height of the melting layer is then estimated by using the standard formula
based on hydrostatic approximation.
2) Mixing ratio at melting level is also obtained by means of interpolation of

retrieved humidity.
3) Mean mixing ratio in the lowest one km is also estimated by using retrieved
specific humidity profile at standard pressure levels.
4) Lapse rate between surface and melting layer is easily estimated by dividing
surface temperature by height of the mixing layer (in km).
These parameters are substituted in the formula for WI given above.
WI knots 0 - - 0.5x 0.5 deg
Frequency: Hourly.
Lat, lon, WI
Domain: 40 S to 40 N, 30 E to 120 E.
15.6.5 Applications: Wind index provides guidance on the maximum possible wind
gusts that can occur with given atmospheric conditions, if convection were to
occur. This is useful for generating short-range warnings and forecasts.
96
16. DATA AVAILABILITY
All derived products from the INSAT-3D/INSAT-3A/Kalpana-1 satellites are made
available to users through IMD website www.imd.gov.in. For more information, please
contact:
A.K.Sharma
Deputy Director General of Meteorology
National Satellite Meteorological Centre
India Meteorological Department
Lodi Road, New Delhi - 110 003 , INDIA.
Phone : +91-11-24697473
Fax: +91-11-24642249
E-mail : ak56.sharma@imd.gov.in
97
98

Insat3d Catalog

Uploaded by

Copyright:

Available Formats

Insat3d Catalog

Uploaded by

Document Information

Original Description:

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

Insat3d Catalog

Uploaded by

Copyright:

Available Formats

0

INSAT-3D DATA PRODUCTS CATALOG

This catalog describes the INSAT-3D payloads, processing system

3. INSAT-3D Payloads ...................................................................... 7

3.2 Atmospheric Sounder ................................................................................................................ 9

3.3 Mode of operation of INSAT-3D, Imager and Sounder: ............................................................... 10

3.4 Data Relay Transponder (DRT): ................................................................................................. 10

4.INSAT-3D Meteorological Data Receiving Earth Station (IMDRES) .. 12

4.2 Quick Look Processing & Archival System .................................................................................. 14

4.3 IMDRES Software Components ................................................................................................. 15

5. INSAT-3D METEOROLOGICAL DATA PROCESSING SYSTEM (IMDPS) 15

5.3 Data Processing System (DPS)................................................................................................... 18

5.4 IMDPS Data Products Generation Software (DPGS) ................................................................... 19

5.5 Data Dissemination Server ....................................................................................................... 23

5.6 Ancillary Data Processing System (ADPS) ................................................................................... 23

5.7 Database Management System ................................................................................................ 24

5.9 Data Dissemination Software (DDS) .......................................................................................... 24

6. Geo-physical Product Generation................................................ 25

8. INSAT-3D Data Products Type and Format ................................... 27

8.2 INSAT-3D Data Products Types ................................................................................................. 27

9. IMDPS Storage Infrastructure ..................................................... 31

10. Network Clients System ........................................................... 33

11. Satellite Image Display System (SIDS) ........................................ 33

12. INSAT-3D IMAGER SAMPLE DATA PRODUCTS ............................. 35

12.2 Imager Sector Images............................................................................................................. 40

13. INSAT-3D IMAGER GEO-PHYSICAL PARAMETERS ........................ 41

13.2. Satellite Rainfall Estimation ................................................................................................... 44

13.3 Operational Implementation of High-Resolution Rainfall Measurements (Hydro-Estimator) ....... 47

13.4 Sea Surface Temperature (SST) .............................................................................................. 49

13.5 Cloud Motion Vectors (CMV) ................................................................................................. 52

13.6 Water Vapor Wind Vector (WVWV)........................................................................................ 56

13. 7 Upper Tropospheric Humidity (UTH) ...................................................................................... 60

13.8 Snow-Cover Mapping ............................................................................................................. 62

13.9 Snow Depth Estimation .......................................................................................................... 65

13.10 FIRE IDENTIFICATION........................................................................................................... 67

13.11 Smoke Identification (SI)....................................................................................................... 69

13.14 Cloud Mask (CMK) .............................................................................................................. 77

14. SOUNDER ATMOSPHERIC PARAMETERS .................................... 79

15. Application parameters from Sounder derived products ............ 83

15.2 Total and Layer Precipitable Water ......................................................................................... 86

15.3 LIFTED INDEX ......................................................................................................................... 90

15.4. DRY MICROBURST INDEX ....................................................................................................... 92

15.5 MAXIMUM VERTICAL THETA-E DIFFERENTIAL .......................................................................... 94

15.6 WIND INDEX .......................................................................................................................... 95

16. DATA AVAILABILITY.................................................................. 97

 To monitor earth’s surface, carryout oceanic observations and its

To accomplish the mission objectives, INSAT-3D has following payloads on board

• Six channel imager

Meteorological payloads are state-of-art and have significant technological

• Imaging in Middle Infrared band to provide night time pictures of low

INSAT-3D adds a new dimension to weather monitoring through its atmospheric

INSAT-3D configuration details

LOCATION 82 deg E Longitude

MASS AT LIFT- 2060 Kg

MISSION LIFE 7.7 years

PAYLOADS Imager, Sounder, Data Relay Transponder and Satellite

The INSAT-3D Imager is an improved design of VHRR/2 (Very High Resolution

Spectral Wave length Ground Quantization IGFOV

MIR 3.80-4.00 4 Km 10 112