BULLETIN OF THE

NATIONAL NATURAL RESOURCES MANAGEMENT SYSTEM

NNRMS (B) - 37

Indian Remote Sensing Satellite Series and Applications: A Saga of 25 years

March 2013

NNRMS
Department of Space Antariksh Bhavan, New BEL Road Bangalore - 560 231 INDIA

Editorial Advisors
Kiran Kumar AS, Director, SAC Dadhwal VK, Director, NRSC Shivakumar SK, Director, ISAC

Editorial Board
Diwakar PG, Director, EOS Raghavaswamy V, Deputy Director (RSA), NRSC Saha SK, Dean (Academics), IIRS Ajai, Group Director, MPSG/EPSA, SAC

Technical Guidance
Shantanu Bhatawdekar, Associate Director (A), EOS

Technical Support and Compilation

Paul MA, Scientist/Engineer SF, EOS Arunachalam A, Scientist/Engineer SF, EOS

For details and inputs, please write to

Director Earth Observations System ISRO Headquarters Antariksh Bhavan New BEL Road Bangalore 560 231 Email: diwakar@isro.gov.in Fax: 91-80-23413806

Published by
P&PR Unit, ISRO Headquarters on behalf of National Natural Resources Management System (NNRMS) ISRO Headquarters Antariksh Bhavan, New BEL Road Bangalore 560 231

Designed by
Imagic Creatives Pvt. Ltd., Bangalore 560 071

Printed at
Brilliant Printers, Bangalore

PREFACE
The experimental satellite missions, Bhaskara-1 in 1979 and Bhaskara-2 in 1981 with TV cameras and 3-band microwave radiometer (SAMIR), the scientic analysis of images sent down by them, were of signicance to move towards operational capabilities for remote sensing in the country. With the launch of IRS 1A on March 17th, 1988 a new era began in the history of operational applications using remote sensing technology. The successful launch of IRS 1A was one of the proudest moments for not only ISRO Scientists, but also for the entire country as we had reached a stage of maturity in realizing our own satellite in orbit that would address the various requirements for managing natural resources of the nation. ISRO has surged forward in a big way after this eventful day and realized a series of earth observation missions with our own launch capabilities of polar satellite launch vehicle. These missions have specically addressed the requirements of land and water management, cartography, ocean applications, urban and rural applications, disaster management, meteorology, weather and climate related studies. ISRO has accomplished several unique missions with diverse capabilities during past 25 years. Spatial resolution of less than 1 m, diversied spectral bands usage, varieties of spacecraft congurations, complex sensors and various other EO technologies have shown that ISRO is second to none in the area of imaging from space. India perhaps has perhaps the best and well-planned constellation of satellites, serving the cause of national and global requirements. The Oceansat-2 with OCM payload & Ku band pencil beam scatterometer and recently launched MeghaTropiques with MADRAS payload puts India in a unique position of global remote sensing community. One of the most complex satellites built and launched by ISRO in the recent past is the RISAT-1 with a versatile Synthetic Aperture Radar (SAR) payload. This has provided an additional dimension for imaging from space in the form of all-weather capability and dawn-dusk imaging. This capability provides a unique advantage for disaster management and crop monitoring in Kharif season, in addition to host of other capability. ISRO and the remote sensing community in the country is in the verge of celebrating 25 years after IRS 1A launch and it is the right time to revisit the past and evolve plans for the future. This special issue of NNRMS Bulletin has made an attempt to bring out articles in different areas of satellite technology and applications on the above lines. I am sure this will prove to be a good technical reading material, presenting diverse elements of earth observation and applications, which would be of interest to many in the science of remote sensing. PG Diwakar Director Earth Observations System

CONTENTS
1 Trends in Spacecraft Technology with Special Reference to IRS Programme
Annadurai M and Venkata Rao M ISRO Satellite Centre, Bangalore

Page No.

1-17

Indian Electro-Optical Payloads for Earth Observation

Samudraiah DRM Space Applications Centre, Ahmedabad

19-26

Evolution and Growth of Sensors, Optics and Thin Films Technologies at Laboratory for Electro-optics Systems (LEOS)
Kamalakar JA Laboratory for Electro-Optic Systems, Bangalore

27-33

Evolution of Microwave Remote Sensors In ISRO

Tapan Misra Space Applications Centre, Ahmedabad

34-57

Agricultural Applications: Evolution during last 25 Years

Parihar JS and Manjunath KR Space Applications Centre, Ahmedabad

58-75

Land Resources Management and Monitoring

Ravisankar T, Sreenivas K, Ravishankar G, Fyzee MA, Sujatha G, Rajiv Kumar, Manoj Raj Saxena, Shyam Sunder B, and Padma Rani G National Remote Sensing Centre, Hyderabad

76-85

Water Resources Management : Evolution of Space Technology Applications

Venkateshwar Rao V, Sharma JR and Raghavaswamy V National Remote Sensing Centre, Hyderabad

86-97

Advances in Satellite Oceanography and Meteorology

Raj Kumar, Rajawat AS, Rashmi Sharma and Kishtawal CM Space Applications Centre, Ahmedabad

98-107

Page No. 9 IRS Data Utilization for Bio-Resources Management

Prakash Chauhan Space Applications Centre, Ahmedabad

108-115

10

Remote Sensing Applications using IRS Data in Geosciences

Vinod Kumar K, John Mathew, Tapas R. Martha, Arindam Guha, Babu Govindha Raj K, Ritwik Majumdar and Priyom Roy National Remote Sensing Centre, Hyderabad

116-122

11

Achievements and Applications for Urban Planning using Data from IRS Satellites
Venugopal Rao, Kamini J, Ramesh B and Raghavaswamy V National Remote Sensing Centre, Hyderabad

123-130

12

Space Technology for Flood Management: 25 Years of IRS

Srinivasa Rao G, Manjusree P, Bhanumurthy V and.Bhatt CM National Remote Sensing Centre, Hyderabad

131-138

13

National Agricultural Drought Assessment and Monitoring System (NADAMS): Genesis and Way Forward
Sesha Sai MVR, Murthy CS, Chandrasekar K and Raghavaswamy V National Remote Sensing Centre, Hyderabad

139-145

14

Capacity Building in Earth Observation Technology and Applications: Achievements and Future Challenges
Team IIRS Indian Institute of Remote Sensing, Dehradun

146-156

TRENDS IN SPACECRAFt TECHNOLOGY WItH SPECIAL REFERENCE tO IRS PROGRAMME

Annadurai M and Venkata Rao M ISRO Satellite Centre, Bangalore 560 017, India Email:madurai@isac.gov.in

Introduction
Space borne remote sensing of earths features and its environment has become an essential requisite for the prosperity and security of human population on this Planet. Reliable, accurate and timely availability of data is the essence of Remote Sensing (RS) programme of any country/agency for effective management of its natural resources, environment and disasters. Over the past four decades, after the launch of Landsat-1 in 1972, several nations have evolved their own RS programmes and launched a number of satellites both in LowEarth Orbits (LEO) and Geo-stationary Orbits (GEO). Commensurate with the ever increasing demands of the users, the technologies have been evolved both in the elds of observation sensors and satellite platforms. A brief summary on the trends in the spacecraft technology with special reference to Indian remote sensing area is presented in this article.

Evolution of IRS Programme

Indian Space Research Organisation (ISRO) realized the potential benets of remote sensing data for various societal applications in 70s and initiated remote sensing activities by launching its rst experimental RS Satellite, Bhaskara-1 in 1979 followed by Bhaskara-2 in 1981. These missions carried optical (TV camera) and microwave (Radiometer) sensors and provided valuable experience in the development and operation of spacecraft technology as well as interpretation / utilization of RS Data. This experience paved the way for the denition of Indian Remote Sensing (IRS) Programme under which the rst operational satellite, IRS-1A was launched on 17 March 1988. From then onwards, series of IRS spacecrafts were launched with enhanced capabilities in payloads and satellite platforms (Table-1). The whole gamut of the activities from the evolution of IRS missions by identifying the user requirements to utilization of data from these missions by user agencies is monitored by National Natural Resources Management System (NNRMS), which is the nodal agency for natural resources management and infrastructure development using remote sensing data in the country. Apart from meeting the general requirements, denition of IRS missions based on specic thematic applications like natural resources monitoring, ocean and atmospheric studies and cartographic applications resulted in the realisation of theme based satellite series, namely, Resourcesat, Oceansat, Cartosat and RISAT.

Development of IRS Spacecraft Technology

Remote sensing is a multi-disciplinary domain and substantial progress has been made in several technologies during the past few decades. As the demand for four

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Resolutions Spatial, Spectral, Radiometric and Temporal increase, the complexity of imaging sensors and in turn the satellite systems, both mechanical and electrical sub-systems, also increases. While reliability calls for heritage systems, lessons learnt and the new requirements warrant new technology developments. A brief overview on these technology developments is given in the following paragraphs. Table 1: List of IRS Satellites launched S.No 1 2 3 4 5 6 7 8 9 10 11 Satellite IRS-1A IRS-1B IRS-P2 IRS-1C IRS-P3 IRS-1D IRS-P4 (Oceansat-1) TES IRS-P6 (Resourcesat-1) IRS-P5 (Cartosat-1) Cartosat-2 Year of launch March 1988 August 1991 October 1994 December 1995 March 1996 September 1997 May 1999 October 2001 October 2003 May 2005 January 2007 S.No 12 13 14 15 16 17 18 19 20 21 Satellite Cartosat-2A IMS-1 RISAT-2 Oceansat-2 Cartosat-2B Resourcesat-2 Youthsat Megha-Tropiques RISAT-1 SARAL Year of launch April 2008 April 2008 April 2009 September 2009 July 2010 April 2011 April 2011 October 2011 April 2012 Feb 2013

Mechanical systems
The three major elements of the mechanical systems of a spacecraft are structure, thermal control and mechanisms. Structure The main function of the spacecraft structure is to safely house all the payloads and subsystems including propulsion and provide adequate shielding against the hostile environment during launch phase and in-orbit phase. Like any other system, realisation of structure goes through design/analysis, fabrication and testing/qualication phases. During design phase, major inputs are launch loads, clearance in the launch vehicle envelope, layout of sub-system package locations, alignment and stability specications for payload and sensors, eld-of-view of sensors and antennae etc,. Dynamic coupling between structure and appendages like solar arrays, scanning systems and large deployable antenna will impact the attitude control and stability, which is addressed in this phase. For example, the natural frequency of deployed solar array was 0.4 0.5 Hz in IRS-1C type bus and the same was increased to 5 Hz in Cartosat-2 bus by additional stiffening to meet the agility requirements. For sub-meter resolution satellites like Cartosat-2C and Cartosat-3, methods of damping high frequency microvibrations due to moving elements (wheels, gyros) need to be addressed. Optimisation of structural mass to minimize its contribution in the overall satellite mass is another important parameter for design and how it was evolved in IRS missions is shown in Figure1. Structural analysis involves modeling of the spacecraft using Finite Element (FE) methods and study/ predict the response of various elements / sub-systems for the quasi-static and dynamic inputs. Subsequently, this model is integrated with the launch vehicle model and goes
Fig. 1: Evolution of structural design

through a coupled load analysis from which

will be derived. The complexity of this FE modeling has increased manifold as shown in Figure1, keeping its pace with the computing power and speed of the computers. Evolution of Various Structures Structure fabrication process involves fabrication of inserts, tooling and jigs/xtures for assembly. The cuboid shaped structure
Fig. 2: Resourcesat payload platform structure

with central load-bearing cylinder and removable equipment panels was evolved for IRS-1A satellite and same conguration was standardized (with improvements) as I 1.5K Bus. In IRS-1C and Resourcesat, a separate Payload Platform (PPL) made of Carbon Fibre Re-inforced Plastic (CFRP) cylinder and payload mounting decks was introduced as shown in Figure 2. Cartosat-2 bus was designed around the payload camera housing with a hexagonal shaped structure as shown in Figure 3. Radar Imaging Satellite (RISAT-1) is the heaviest satellite (1858 kg) built in the IRS series so far, where the mainframe platform was attached to the SAR payload structure as shown in Figure 4. Recently, two small buses for Indian Mini Satellites (IMS-1 & IMS-2) were developed with 100 kg and 400 kg mass capability respectively (Figure 4a). Fabrication & Testing While aluminum honeycomb is extensively used for panels / decks, CFRP material is used for main cylinder and payload platform. Due to its low co-efficient of thermal expansion and high stiffness to mass ratio, CFRP is mainly used for payload camera structure for Cartosat type spacecraft. During testing phase, the structural sub-assemblies go through Non-Destructive Tests (NDT) like holographic techniques as well as tests like bond-pull of inserts to ensure the quality and workmanship aspects. In case of a new

Fig. 3: Cartosat-2 structure

structure being developed rst time for a

N N R M S

B U L L E T I N

selection of materials/pre-pregs, adhesives,

M A R C H

2 0 1 3

the expected launch loads to the spacecraft

programme (like in IRS-1A, Cartosat-2), a qualication model goes through a static loading test to establish the safety margins in design. Finally, the structure is qualied by subjecting the fully assembled spacecraft to sine and low level random vibration as well as acoustic tests simulating the launch environmental loads. Thermal Control System Thermal Control System (TCS) has to maintain the temperature of all sub-systems and payload elements within their specied limits during the mission life. This is normally achieved through heat balancing between external input loads (due to sun, earths albedo) and internal loads (due to power dissipation of sub-systems) and by employing several passive and active techniques. TCS design should also address extreme temperature conditions which can occur under contingencies like safe mode and power emergency mode and accordingly plan/implement corrective measures like automated thermal load management and contingency heaters etc,. Major inputs for an effective design of TCS include orbital parameters like sun aspect angle variation, eclipse duration & satellite orientation in different modes, duty cycle of payload operation, power dissipation of sub-systems etc,. Temperature limits of all sub-systems including any critical requirements, materials used in the subsystem packages and their physical properties
Fig. 4: RISAT-1 structure with SAR antenna

also need to be spelt out at design phase. TCS design and analysis involves generation of a

mathematical model of iso-thermal nodes for sub-system packages/appendages and study the interactions between them. The number of nodes increases with the complexity of satellite and as the limits of temperature control become narrower and stringent, like electro-optical payload cameras, battery etc,. While the number of these nodes was < 50 in Bhaskara and ~ 450 in IRS-1A , this number went up to 50,000 in Oceansat-2. The wide variation in the local time of satellite pass (6AM/6PM for RISAT & SARAL, 9.30 AM for Cartosat-2, 10.30 AM for Resourcesat and 12.00 Noon for Oceansat) and orientation of satellite in different modes (like sun pointing during non-payload operation for Cartosat series, off-nadir viewing during payload operation in Cartosat and RISAT) offer challenges in the TCS design which were successfully met. Some of the critical thermal requirements implemented in IRS satellites include

 SWIR CCDs - -10 deg.C in IRS-1C Star sensor CCD - -20 deg.C in IRS-1A/1B Ni-Cd Battery - 5 +/- 5 deg.C With the increased usage of high density Very Large Scale Integrated (VLSI) circuit devices to miniaturize the electronics packaging, TCS design has to take care of heat transfer paths at device, PCB and package level so that their case / Junction temperatures are within the limits. Implementation of TCS involves usage of passive elements like high & low emissive tapes, paints, Optical Solar Reectors (OSR) and multi layer insulation blankets, radiator plates etc., and active elements like heaters. Heat pipes in conjunction with copper braids were employed in Resourcesat to control the temperature of 12 K element CCDs of LISS-4 camera. Several new processes like black painting on kapton tape, black anodisation of aluminum, micro arc oxidation of magnesium alloy, hard anodisation, integral black anodizing on magnesium alloys , conductive chromating on magnesium alloys etc,. have been developed and qualied. With the introduction of microprocessor based Bus Management Unit (BMU) in Cartosat-2, autonomy in thermal management in terms of auto temperature controllers was
Fig. 4a: IMS-2 (SARAL) structure

realized with programmable upper & lower temperature limits. As an off-shoot of the

evolution in TCS, several materials like thermal paints, kapton tapes, rigid OSR, precision thermistors, tape /foil heaters and heat pipes etc., have been indigenously developed. Validation of TCS design by testing is very important particularly when new congurations are introduced. For newly designed electronics packages, thermal performance is veried at bread-board level in a smaller thermo-vacuum chamber and any hot spot devices are identied. Performance of overall thermal design is veried during the thermovac test on the spacecraft by simulating the external loads using tape heaters pasted on the outer surfaces or using Infrared lamps placed near the spacecraft inside the TV chamber. Thermal balance test was conducted at CNES facility at Tolouse for IRS-1A and same was conducted in the 9 m dia Large Solar Simulation Chamber (LSSC) at ISAC on the IRS-1C electrical cum thermal model spacecraft. Since this bus is standardized for subsequent IRS spacecraft, this test was conducted on a limited scale on new payloads like Resourcesat-1, Megha-tropiques and SAR antenna tile in RISAT-1. Mechanisms Mechanisms in IRS spacecraft mainly consists of in-orbit deployment and scanning/rotation of Solar arrays Hold-down release of payload tilt/steering & scan mechanisms Payload antennae 5

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Payload cameras - 20 +/- 2 deg.C with a gradient of < 2 deg.C

Since successful deployment and operation of these elements is mission critical, achieving high reliability is utmost important from the design phase. Quasi-static and dynamic loads during launch, physical properties of the elements to be deployed (mass, CG, MI), expected shock loads at deployment etc,. are some of the critical inputs for design. Selection of materials including dry lubrication coatings like Molybdenum Sulphide (MoS) and fabrication of mechanical components/assemblies to a high precision need special attention. In IRS spacecraft, solar arrays are stowed during launch and deployed automatically immediately after satellite separation from the launch vehicle using onboard timers. The size of the solar array was increased from 9.2 m2 in IRS-1A to 15.1 m2 in IRS-1C and same size is retained for later missions with IR-1.5K bus. The accordion type solar array deployment with Closed Control Loop (CCL) and spring loaded design has proved to be very reliable with 100 % successful results. Deployment and payload hold down mechanism was implemented i n I R S - 1 C / 1 D f o r t h e PA N c a m e r a and in Oceansat-1&2 for the Ocean Colour
Fig. 5: SAR Antenna deployment (RISAT-1)

Monitor. Later this was extended to release of LISS-4 rotating deck in Resourcesat-1&2 with a overall mass of ~ 250 kg. Meeting the requirements of accurate pointing of camera optical axis before and after deployment, maintaining the critical gap of 2-3 mm after deployment at hold-down points for free rotation of camera are some of the challenges in these deployments. Deployment of the Synthetic Aperture Radar (SAR) antenna in the RISAT-1 spacecraft is a major achievement. With a total mass of ~ 900 kg and size of 2 x 6 m2 (Figure 5) design, development and qualification of this antenna deployment went through many critical phases. Against the stringent requirement of co-planarity of < 2 mm among all the tiles after deployment, 0.5 mm was achieved. Other antenna deployments realized in IRS missions are holddown release mechanism of the Scatterometer antenna in Oceansat-2 and dual gimbal

Fig. 6: Dual gmbal antenna mechanism (Cartosat-2)

antenna (Figure 6) in Cartosat-2/2A/2B.

cable cutter for SARAL and Shaped Memory Alloy (SMA) based pin-puller for Cartosat-2 series. New challenges envisaged for future missions are development of larger diameter unfurlable antenna for L- band SAR and satellite docking mechanism that may enable spacecraft in-orbit servicing and life extension.

Electrical Systems
Electrical systems in IRS satellite are power, base-band and RF systems, attitude and orbit control system including sensors, control electronics and actuators, inertial systems. Power System Main functions of the power system of any spacecraft are generation of power, storage, regulation / conditioning and distribution of power supply to payload and sub-systems throughout its mission life. All the IRS spacecraft being Low Earth Orbiting (LEO) missions, power system design has to cater to the large number of thermal cycles (~ 25,000 for 5 years life) of about -100 to + 100 deg.C and the eclipse duration of about 35 minutes in every orbit. All the three elements of the power system solar arrays for generation, chemical batteries for storage and power electronics for power conditioning and supply have been evolved and improved with the growing requirements of IRS satellites. Solar Array Majority of IRS spacecraft used silicon solar cells for power generation since IRS-1A with improvements in efciency (upto 13 %), usage of large size cells (4 x 6 cm2) in later missions. However, Gallium Arsenide based Triple junction / multi junction cells with an average efficiency of 27 % are used in Cartosat-2 series and RISAT satellites to minimize the solar array mass for high platform agility. Evolution of the power requirements and achieved power to weight ratio in IRS missions is shown in Figure 7. In the area of solar panel fabrication, several improvements like usage of CFRP substrate (since Resourcesat-1), use of silver plated copper bus bars and inter-connects
Fig. 7: Evolution of power generation in IRS

(Cartosat-2 & Oceansat-2) with welding process etc,. have been evolved.

Battery For energy storage, Nickel-Cadmium batteries have been used in a majority of IRS spacecraft from IRS-1A except in RISAT where Nickel-Hydrogen batteries were used as shown in Figure 8. Under controlled temperature environment of 0 - 10 0C, they can withstand a large number of charge/ discharge cycles, provide high depth of discharge (~ 20%) and long life. Though imported battery cells were used in many IRS satellites, indigenous Ni-Cd cells developed by Vikram Sarabhai Space Centre (VSSC) have been used in the TES spacecraft. To meet the peak power requirements and also due to obsolescence of Ni-Cd batteries, Lithium-Ion batteries are being used in the recent IRS and small satellites (SARAL, Cartosat-2C). Their main advantage is very high depth of discharge (~ 35 %) besides lighter 7

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

New technologies developed for deployment mechanisms include paran wax actuated

weight and wider temperature range of operation (5 20oC). Power Electronics Main functions of power electronics include regulating the generated power, controlling the charge and discharge current, distributing power meeting user requirements, providing fault tolerance and isolation against any malfunction, and monitoring of the power generator and storage elements. In IRS-1A design, distributed bus concept was evolved with two battery tied raw buses provided to various sub-systems. In IRSFig. 8: Evolution of power storage in IRS

1A/1B, a 28V regulated bus was generated using sequential switching shunt regulator

in conjunction with a shunt resistor package and battery discharge regulator. This regulated bus was used to operate relays and supply to selected sub-systems like heaters, wheels, TWTA etc,. This design was changed from IRS-1C to Pulse Width Modulated (PWM) switching scheme of solar array strings where two (out of total 10) strings connected directly to battery, are operated in duty cycle mode and rest of the strings are switched ON or OFF depending on the bus load using taper charge regulators. The 28V regulated bus was discontinued and only two raw bus lines of 28 42V with fuse protection is distributed to all sub-systems using modular type package design. To handle the high power (~ 4KW peak) in RISAT-1, new power design with 70 AH Nickel-Hydrogen batteries, Battery Discharge Regulators (BDR) and SAR were implemented. While the SAR payload and some sub-systems were operated with 70V bus, other satellite systems were operated at 42V bus. Battery emergency logic evolved in IRS-1A to isolate faulty power bus and share the load with bus parallel relay is continued in all IRS spacecraft and another Soft Emergency logic was included in the recent spacecraft where load shedding is done if the bus voltage reaches a pre-dened set value. Power supply to sub-systems is done using DC-DC converters with raw bus input and different voltage outputs. Higher efciency, low noise / ripple, small in size and mass and compatible to EMI/EMC standards are the criteria for evolving new topologies of the DC-DC converters. In IRS spacecraft, most stringent requirements from payload system less than 1 mV ripple to meet the SNR specications have been met using push-pull topology converters. DC-DC converters with efciency of ~ 80 % could be achieved and Hybrid Micro-Circuit (HMC) packaged high power (30 W) converters have been developed. Base - Band Digital Systems The main functions of base-band digital systems in IRS Spacecraft may be classied into i) low bit rate Telemetry, Telecommand, ii) high bit rate payload data handling, and iii) on-board data storage systems Telemetry Telemetry (TM) system collects the House keeping parameters of on-board systems and transmit the same to ground stations through TM transmission to continuously monitor the health of the satellite. TM system used in IRS spacecraft was evolved technologically to provide better reliability, exibility and growth in performance, both in terms of number of channels (715 in IRS-1A to 7048 in RISAT-1) and sampling rates (32 m sec in IRS-1A to 2 m sec in Cartosat-2). New features like multiple formats, distributed multiplexing scheme, increased storage capacity with variable sampling rate were incorporated. Telemetry data rate was also increased from 256 bps in IRS-1A to 4 Kbps in Cartosat-2 with PSK

was own in small satellites (IMS-1, SARAL) and can be adopted in future IRS satellites. Telecommand Telecommand (TC) system controls the satellite and payload operations by receiving the necessary ground commands. Though the basic design of TC based on the BCH code was evolved in IRS-1A, several new features were incorporated in the following IRS satellites. With the increased command requirements for global coverage and satellite maneuvers required for payload operations was evolved in Cartosat-2. Also, a Payload programming software based on-board autonomy was implemented in Cartosat-2, Resourcesat-2, RISAT-1 and Megha-tropiques etc,. Features like on-board timer (OBT) linked time tagged commands, event based commands, congurable command blocks, checksum logic and variable pulse width commands etc,. enabled improved and exible mission operations. CCSDS format based command system implemented in the SARAL satellite could be adopted in future IRS missions. However the future trend will be to have more and more autonomy on satellite operations that reduces the need for frequent ground command requirements. Baseband Data Handling Base-band Data Handling (BDH) system receives payload data, formats it after adding auxiliary information (attitude & orbit, on-board time, line count and HK data etc.) and modulates it on the RF carrier for transmission or stores the data in the on-board storage for later transmission. In pace with the developments in the payloads, BDH designs were evolved to handle high data rates, multiple formats / interfaces, data compression and pre- processing etc,. Figure 9 shows the growth in the data rates of IRS Missions from IRS-1A. Data compression techniques like JPEG-2000, DCT based Wavelet compression etc, have been implemented in Cartosat-1 and Cartosat-2 series. In Resourcesat-1, bit sliding to select 7 out of 10 bits and pixel sliding to select 4K out of 12K CCD pixels in LISS-4 were implemented in BDH system. In addition to standard TTL/ECL interfaces in earlier IRS satellites, MILStd-1553B interface in Oceansat-2, Megha-tropiques and LVDS with serialiser-deserialiser interface in RISAT have been implemented. High frequency operation maintaining the clock stability, signal integrity and meeting the BER specications were some of the challenges in the BDH system of IRS missions. Future trend in the data handling may require a high level of on-board data processing such that the data product is delivered to the user directly. On-board Data Storage On-board data storage requirement in IRS satellites came due to two reasons to provide global coverage of payload operation and/or the payload generated data is higher than real time transmission rates. With the experience of tape recorders own in IRS1C/1D, solid state memory based recorders were selected from Resourcesat-1 and Cartosat-1 onwards. Commensurate with payload data rates, the storage capacity has increased as shown in Figure 9. With the
Fig. 9: Evolution of data handling systems

advent of high density memory modules

N N R M S

B U L L E T I N

(maximum ~3000 commands in a day), high bit rate (4Kbps) command system with PSK modulation

M A R C H

2 0 1 3

modulation. Packet telemetry with Consulative Committee for Space Data Systems (CCSDS) format

like DRAM and SDRAM (currently 2 GB each module), modular designs were adopted for Solid State Recorder (SSR) memory boards (Figure 10) while mission specic software and operational modes are incorporated in the FPGA based controller. As the SSR operation has become mission critical in IRS missions like Oceansat-2, RISAT and Megha-tropiques etc., high reliability is the essence of their design.

RF Systems
The radio frequency systems of IRS spacecraft encompasses i) Telemetry, tracking
Fig. 10: 50Gb Solid State Recorder (SSR) board

and Command (TTC) system, ii) payload data transmitter system and iii) Global Positioning

Systems (GPS) receiver based satellite positioning system. TTC System The Telemetry, Tracking and Command (TTC) system own in IRS satellites is based on the S-band coherent transponder (back-up VHF commanding was last own in IRS-1A/1B). TTC system consists of three elements a transmitter for telemetry data and ranging tones down link, a receiver for telecommand and ranging tones uplink, and a local oscillator. These systems were designed using discrete components in IRS-1A. Due to obsolescence of many RF components, miniaturized design using HMCs, surface mount devices, band pass lters and oscillators was evolved and own in IRS-P5. Later, to meet the demand for high bit rate telecommand (4 Kbps) , the modulation scheme was changed from FSK/FM to PSK using costas loop based demodulator and was rst own in Cartosat-2. For small satellites, non-coherent TTC transponder was developed without ranging function. An FPGA based miniature digital receiver was developed for SARAL satellite with CCSDS based command format. Data Transmitter The data transmitter receives the formatted digital data stream from BDH system, modulates it on the RF carrier either in BPSK or QPSK mode and after power amplication transmits to the ground station. As per the frequency allocation regulations, either S-band (2-4 GHz) and/or X-band (8 GHz) was used in all IRS Missions. Design improvements like PLL based circuits, usage of GaAs FETs, miniaturization of packages, integrated power modules have been realized. In addition to the 20W / 40W Traveling Wave Tube Ampliers (TWTA) used in IRS-1A/1B, 1C/1D, IRS-P5/P6 Satellites, 10W Solid State Power Ampliers (SSPA) were used for moderate data rate missions like Oceansat-2 and Meghatropiques. While
Fig. 11: Phased Array Antenna (RISAT-1)

shaped beam conventional antennae were

Antenna (DGA) (Figure 11) were indigenously developed for Cartosat-2 series and RISAT-1, where Satellite is to be maneuvered for off-nadir payload operation and to point the transmitted beam to the ground station. Future missions like Cartosat-3 envisage Ka-band data transmitter to take care of the high data rate requirement. Satellite Positioning System (SPS) SPS is an on-board GPS receiver which the host satellites position and velocity are computed. An 8-channel SPS was rst flown in IRS-P4 spacecraft and improved versions in all subsequent satellites. SPS operates in the L1 (1575.42 MHz) frequency of GPS satellites and with C/A code. While the RF front end including the patch antenna, Low noise amplier are realized by CMG/ISAC, the Receiver Core Engine (RCE) was realized through industry. To take care of the radiation induced single event upsets, new RCE design with Rad-hard FPGAs and 10 channel with a back-up of 8 channel receiver based SPS was own from Cartosat-2B onwards (Figure 12). SPS data interface with MIL-Std1553B bus was introduced in Cartosat-1 and it was included in the on-board computation of state vector using the GPS based On-board Orbit Determination Software (GOODS). Compared to the Orbit Determination (OD) accuracy of ~ 400 m achieved with S-band tracking in earlier IRS spacecraft, OD accuracy of ~ 10 m has been achieved with SPS. In future IRS missions, development of dual frequency SPS with ionospheric correction will give better than 1 m accuracy. which measures the pseudo-ranges from

Attitude and Orbit Control System (AOCS)

The main functions of the AOCS are maintain the attitude of the spacecraft in the desired orientation and platform stability within the specied limits and maintain the satellite in its intended orbit throughout its mission life. These functions are achieved by AOCS with
Fig. 12: SPS antenna & RCE electronics

three major elements attitude sensors, AOC

11

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

used in a majority of IRS missions for data transmission, Phased Array Antenna (PAA) and Dual Gimbal

electronics and control actuators which operate in a closed loop. A brief summary of these elements used in IRS satellites is given below : Attitude Sensors These sensors are used to measure the attitude errors of the satellite with reference to celestial bodies like earth, star and sun and the satellite body rate using gyroscopes. In IRS-1A/1B, two types of earth sensors were own to measure roll and pitch errors , namely Static Earth Sensor (SES) and Conical Scanning Earth Sensor (CSES) and in later missions, only CSES was own as a standard bus conguration. A miniature version of CSES using FPGA, built-in DC-DC converter was developed and own in Resourcesat-2. These earth sensors could provide roll and pitch errors to an accuracy of +/- 0.07 deg,. with on-board correction of systematic errors. To measure the Yaw error, two sensors Analog Yaw Sensor (ANYS) and a CCD based Precision yaw sensor were own in IRS-1A/1B. These sensors were replaced by High accuracy Digital Sun Sensors (DSS) from IRS-1C onwards. A coded optical mask along with a pin hole and silicon photo-diode behind the mask measures the sun aspect angle to give spacecraft Yaw error near Poles with a large cross eld-of-view. An
Fig. 13: Sensors own in IRS-1A

on-axis sun sensor with narrow FOV is used to validate the error value before update. Various sensors own in IRS-1A Satellite are shown in Figure 13. Star sensor with a 2K Pixel linear CCD array was rst own in IRS-1A/1B with a eld-of-view of 7.5 deg. In IRS-1C/1D, the design was modied with an area array CCD of 288 x 384 pixels covering an FOV of 6o x 8o. These CCDs were cooled to -20 deg.C using thermo-electric coolers. Based on these experiences, several improvements were implemented in the later star sensors like better baffle design for stray light / background illumination, integrated optical head and processing electronics housing, algorithms for quaternion based attitude with Kalman ltering, MIL-Std-1553B interface with AOCE etc,. In Cartosat-1 and Resourcesat-1, a new design with Multi-Pinned-Phase (MPP)

Fig. 14: Star sensor (Mk-1)

CCD of 1K x 1K pixels was evolved covering a FOV of 20o x 20o. This design has considerably reduced the dark current thereby improving the sensitivity and eliminating the requirements of thermo-electric coolers. Star sensor mechanical design was also improved to have better alignment stability, thermal control and radiation protection etc. An attitude determination accuracy of < 20 arc.sec could be achieved with star sensor Mk-I in Cartosat-2 and Resourcesat-1/2

accuracies of the order of < 10 arc.sec for all axes in future missions like Cartosat-3, star sensor Mark-III with dual head and single processor is being developed. Other sensors own in IRS satellites include, tri-axial magneto meters to measure about the satellite axes, which is used in the momentum dumping of reaction wheels; four-Pi steradian sun sensors for initial sun acquisition and safe mode, solar panel sun sensors to orient solar panels towards sun. Dynamically Tuned Gyroscope (DTG) based inertial reference unit , shown in Figure
Fig. 15: Inertial reference unit

15, is used in all IRS Spacecraft to accurately measure the angular rates about all three axes in the inertial reference frame. Compared to the DTG own in IRS-1A/1B with separate electronics package, many improvements were incorporated in the later missions like temperature control of the DTG cluster, ne rate measurement circuits, micro-controller based electronics, MIL-Std-1553B interface with AOCE, built-in DC-DC converters etc,. Also, certain process improvements in fabrication of cluster, exures etc., resulted in achieving the drift and scale factor stability by an order of magnitude (~ 0.005 deg/ hr against spec of 0.05 deg/hr in drift and < 50 ppm against Spec. of < 500 ppm in scale factor, respectively). A miniaturised DTG was developed and own in SARAL (see Fig.16). ber optic gyros with no moving elements are being developed for future Cartosat series to

Fig. 16: Mini-DTG

reduce the noise performance. Attitude and Orbit Control Electronics (AOCE) AOCE receives the attitude errors from sensors and drives the actuators according to the control algorithms and gains. AOCE design in IRS-1A/1B was based on hardware logics using discrete devices and Harris 1802 processor. In IRS-1C/1D, software based AOCE using 80C86 micro-processor was implemented which was later improved in Cartosat-1 / Resourcesat-1 using Marconi 281 microprocessor and MIL-Std-1553B bus interface was introduced. An integrated Bus Management Unit (BMU) encompassing the functions of AOCE, telemetry & telecommand, sensor signal processing, 13

N N R M S

B U L L E T I N

the components of earths magnetic eld

M A R C H

2 0 1 3

spacecraft (Figure 14). To meet higher attitude

Distributed Electronics (IRS-1A)

thermal management with temperature sensors and heaters was rst introduced in Cartosat-2 spacecraft using Dynex 31750 micro-Processor, Actel FPGAs, and ASICs in conjunction with Ada S/W language. In all these spacecraft, hardware based safe mode detection logic is implemented to take care of any attitude loss and orient the spacecraft to Sun pointing mode. In RISAT-1, the BMU was further optimized into an OnBoard Computer (OBC) by replacing FPGAs with ASICs and improving the inter-package harness design. For small satellite bus like

Bus Management OBC (RISAT-1)

SARAL, a miniaturised BMU with two PCBs was developed. Evolution of AOCE coupled with Bus electronics is shown in Figure 17. Commensurate with the demands of satellite maneuvers for payload operation, high pointing accuracy and low drift rates, the control dynamics has evolved new modes of operations and algorithms from IRS-1A onwards. Direct earth acquisition after launch with Inertial Attitude Control (IAC) mode, dynamic friction compensation for wheels, estimation and compensation of gyro drift, continuous dumping of wheel momentum using International Geo-magnetic Reference Frame (IGRF) model, orbit maneuver through sequencer etc., have been implemented. For Cartosat-2 series, step-and-stare operation of payload camera requires acquisition of ground target area to be imaged and continuous rotation of spacecraft with controlled body rate demand intense mathematical computations on-board, which were successfully achieved by AOCE. Control Actuators The control actuators used in IRS spacecraft are magnetic torquer coils, reaction wheels and reaction control system with thrusters. Magnetic torquers are mainly used for momentum dumping of the reaction

Miniaturised BMU (RISAT-1)

Fig. 17: Evolution of on-board bus electronics

wheels whenever their speed exceeds the set

increased from 20 Am2 in IRS-1A/1C to 60 Am2 in Cartosat-1 onwards. Bipolar operation of magnetic torquers was implemented in Cartosat-2. Reaction wheels are used for ner control of attitude in the mission phase as well operation or orbit corrections. In IRS-1A to 1D satellites, three wheels were mounted along the satellites axes and fourth wheel was skew mounted making equal angle (54.7o) with all three axes. From Resourcesat-1, the four wheels are mounted in tetrahedral
Fig. 18: Reaction wheels

configuration. Though both mounting congurations give zero momentum, later conguration gives higher resultant torque. The capacity of RWs in terms of angular momentum and reaction torque has steadily gone up from 5 & 10 NMS/ 0.05NM in IRS-1A to 50 NMS/0.3 NM in RISAT-1 to meet the mission requirements (Figure 18). Parameters that effect the life of the wheels like lubricant quantity & its friction, ball bearings, oil cage stability etc., have been optimized during the evolution of the IRS missions. The wheel drive electronics housed in a separate package in IRS-1A was later modularized with commutation electronics (using Hall effect sensors) mounted inside the

Fig. 19: Mini reaction wheels (0.36NMS)

y wheel unit. Bearing and casing heaters were incorporated in the wheel to reduce the viscosity/friction of the lubricant oil if required. Mini reaction wheels of 0.36 NMS have been developed and own in Youthsat (IMS-1) as shown in Figure 19. To meet the high agility requirements in future missions like Cartosat-3, control moment gyros with high torque capability are being developed. Inertial Systems Other inertial systems used in IRS spacecraft are - Solar Array Drive Assembly (SADA), Antenna Scan Motors (ASM), Payload Steering Mechanism (PSM) and Payload Scan Mechanism (PSM). SADA system is designed with a stepper motor and slip ring assembly for power transfer. Each solar panel array is attached to SADA through a yoke and the power and signal cables (from Solar Panel (SP) sensors and temperature sensors) are routed through slip rings. To meet the increased solar array size / moment of inertia, the motor torque was increased from 0.85 NM in IRS-1A to 4 NM in Resourcesat-1. The SADA Drive electronics design was improved by incorporating PROM based logics like auto-capture of sun after eclipse exit, variable torque modes, failure detection and auto change 15

N N R M S

B U L L E T I N

as large angle maneuvers either for payload

M A R C H

2 0 1 3

limit. Capacity of the magnetic torquers was

over etc. Micro-stepping of motor drive was introduced in Resourcesat-1/Cartosat-1 to reduce the attitude disturbance of platform. Potentiometer based angular position of solar array and open loop drive with periodical updates from sun sensor are other improvements in SADA. A unied SADA design (common for IRS & GEOSAT) where the SADA motor is driven by BMU/OBC as per mission required prole was recently own in RISAT-1. Antenna Scan Mechanism motors are used in microwave payloads of IRS missions. These scan motors were own for the Multi-frequency Scanning Microwave Radiometer (MSMR) in Oceansat-1 (IRS-P4) , Scatterometer antenna scan mechanism in Oceansat-2 and MADRAS scan mechanism in Meghatropiques. Stability of rotating speed with feedback loop, precise angular position measurement using optical encoders, and electro-mechanical design for long life are critical requirements for these systems. Payload Steering Mechanism was used for tilting the PAN camera in IRS-1C/1D and LISS-4 camera deck in Resourcesat-1/2 by +/- 26 deg. across the satellite ground track for faster revisit. The PSM was designed with a stepper motor and an absolute resolver with coarse and ne angle measurements. Pop-up deection of the PSM after hold-down release by 2-3 mm using a corrugated diaphragm was a novel design implemented in these mechanisms. Reaction Control System (RCS) RCS used in IRS satellites is a mono-propellant based system with Hydrazine (N2O4) as the fuel and gaseous Nitrogen (GN2) as pressurant and operating in blow-down mode. In IRS-1A/1B, RCS was congured with 16 nos. 1N Thrusters and 4 fuel tanks each with 21 L capacity. A combination 1N and 11N thrusters was implemented in IRS-1C/1D and follow-on satellites. Mounting of all thrusters in the Bottom deck (with canting of 30o for 1N) was evolved from Resourcesat-1. Design of Fuel tank was changed from elastomeric bladder in IRS-1A to surface tension type with Propellant Management Device. Fuel tanks with a maximum capacity of 390 L was qualied and own in Resourcesat-1,2, Oceansat-2 and Megha-tropiques and minimum capacity of 3.5 L was developed for IMS-1. The Neoprene rubber seats in the Flow control valves have been changed to Teon since RISAT-1. All the components / materials of the RCS including Titanium tank forgings have been developed indigenously except thruster bed heaters and Shell-405 catalyst material.

Assembly, Integration & Testing (AIT)

The AIT activities spread across the realisation of the spacecraft starting from conguration evolution to launch phase. In a complex spacecraft like IRS, close interaction between the three teams of AIT, namely, Mechanical Integration, Electrical Integration and Ground Check Out (GCO) is crucial for successful operations. With the increased complexity of payload and mainframe sub-systems in IRS missions, improvements in mechanical integration activities like package layout preparation with computer aided techniques, 3D simulation for FOV studies, accurate alignment measurements with photogrammetric techniques, precise estimation & measurement of satellite physical parameters and safety measures in handling and transportation etc., have been realized. Similarly, in the area of electrical integration, improvements in the grounding schemes, compliance with EMI/EMC specications, optimisation of harness routing, RF radiation measurements in anechoic chambers, computerized data base for electrical interface details etc., have been implemented. Test set-ups for special tests like interferometric Modulation Transfer Function (MTF) measurements, SNR and vignetting measurements on optical payloads have been established during Cartosat-2. Contamination control with clean air chambers are being implemented for ASTROSAT and future optical payloads. The main goal of ground checkout has been standardization, miniaturization, indigenization and automation. GCO activities have gone through signicant improvements since IRS-1A time in terms of high speed computers for data acquisition and processing with client-server architecture, automated test procedures with in-house developed checkout command language, high bit rate payload data acquisition and processing (upto 160 Mbps), remote control of solar array, battery simulators and other stimuli units, built-in safety alarms, ber-optic links for data transfer during

external industry was a signicant achievement saving foreign exchange.

R&QA aspects in spacecraft programmes is a very signicant element and encompasses almost all activities of satellite realization concept to completion. Meeting mission goals of satellites is always dictated by well orchestrated assurance programme and the real impetus for this got dened during onset of operational remote sensing projects. R&QA methods commensurate with space and military design analysis as well as comprehensive non-conformance management. With the increased complexity of payloads and electrical and mechanical design of satellite sub-systems, increased life expectancy, un-interrupted service demands coupled with ever changing technology and its obsolescence, assurance methods also got evolved. Newer techniques of circuit analysis, reliability estimates through Failure Modes Effect and Criticality Analysis (FMECA), Worst Case Circuit Analysis (WCCA), verication for electronic component stress, its de-rating and immunity for space radiation environment have become more critical. Evaluation/characterisation of new materials, life cycle qualication of new processes, adoption of non-invasive quality control techniques, environmental stress screening etc., are introduced. R&QA perspective is also part of system engineering with study of optimal redundancy, fault tolerant design requirements foolproof autonomy features etc. Commercial competitiveness and out-sourcing of hardware realization to industry has added new dimensions to assurance practices. Vendor quality management with several key issues like training and certication of manpower, periodical auditing of vendors facilities and / fabrication and test reports are getting importance. In the recent IRS satellites, scope of software QA has gone up and procedures like Software Requirement Specication (SRS) review, algorithm design review and code-walk-through have been evolved to support this new design entrant. Suitable provisions/protocols/ISRO level standards have been generated. In the area of electronic components, identication of new reliable components with shrinking dimensions, ambitious detectors and high-speed data handling devices have increased component qualication requirements manifold. Replacing obsolete components / to cater to new designs, screening / characterisation of components, design and development of complex hybrid micro-circuits are the challenges in this area. Overall, since the initial days of operational remote sensing satellites, R&QA aspects have continuously evolved in tune with technology, mission goals and global market industry standards were implemented for parts, materials, fabrication processes, test and evaluation,

Conclusion
Since the launch of the rst operational satellite, IRS-1A, the spacecraft technology under IRS programme has been evolved in multi-faceted disciplines to meet the growing user demands for new payloads and services. A brief overview on these developments is given in this article. Contribution from various sub-system groups spread across ISRO Centers/Units in these developments is gratefully acknowledged.

References
Kasturi Rangan, K. (1991). Evolution of Spacecraft Technology in ISRO. Journal of Spacecraft Technology, (ISSN 0971-1600), vol. 1, no. 1, p. 1-29. Conguration Data books of IRS Satellites.

17

N N R M S

B U L L E T I N

M A R C H

Reliability & Quality Assurance Aspects (R&OA)

2 0 1 3

spacecraft thermovac tests etc,. Indigenous development of in-house designed GCO units through

Introduction
Satellite based remote sensing in our country took a soaring start under the agship of Indian Remote Sensing (IRS) programme with launch of IRS-1A satellite in 1988, which later grew into formidable combination of multitude of sensor systems and platforms to provide sustainable data services for innumerable resource management applications such as agriculture, forestry, water resources, town planning, disaster management, sheries etc. Concurrently, several payloads were realized and own for meteorological missions under INSAT series. All the electro-optical payloads own on these missions had performed exceedingly well and had met various user demands. Advancements in the electro-optical payload technology in the last 25 years have not only provided data for operational applications, but also enabled research and development on wide range of applications. Sensor development team at Space Application Centre has relentlessly contributed in the design, development, integration and testing of state-of-the-art electro-optical sensor systems for all these missions. One of the reasons for the success saga of IRS in the 25 years was due to the leapfrogging progress made in the eld of payload technology and this experience paved way for taking up the development of even complex and advanced remote sensing systems in the near future. This review article is an attempt to narrate the evolution of electro-optical payload technology in the last 25 years, highlighting challenges faced during design and realization of all these payloads and also discuss the future prospects for electro-optical payloads.

Evolution of Indian Electro-Optical observation

sensors for earth

Electro-optical payload technology development for remote sensing in our country dates back to 1970s when number of initiatives were taken up like balloon and aircraft based imaging using photographic cameras and thermal scanner and development of six channel Multispectral Scanner (MSS). The development of MSS, helped not only in understanding the nitty-gritty of aircraft electro-optical sensor technology but also the development of image generation, interpretation and denition of spectral channels for IRS. The rst space borne experimental earth observation satellite, christened as Bhaskara-1, was launched in the 1979 carrying Super Vidicon TV cameras (area detector) operating in two spectral bands and provided contiguous pictures with resolution of 1 km covering a swath of 400 km.

N N R M S

B U L L E T I N

Samudraiah DRM Space Applications Centre, Ahmedabad 380 115, India Email: drms@sac.isro.gov.in

M A R C H

INDIAN ELECTRO-OPTICAL PAYLOADS FOR EARTH OBSERVATION

2 0 1 3

The expertise and experiences gained in the development and on-orbit operation of Bhaskara-1&2 payloads provided a strong learning base for the team for taking up design of operational space borne electro-optical payloads. The sensor development team took early initiatives to learn the advanced technologies like the then upcoming Charge Coupled Device (CCD) imaging and realised a single band camera. It was aircraft ight tested along with MSS. The CCD imagery was very impressive (The same could be seen on most of the IRS-1A payload documents). The denition of payloads for IRS-1A was historic. It started with MSS as main payload and CCD Payload as piggyback. Subsequently, the payload team made very detailed assessment and a decision was taken to have IRS-1A with all solid state CCD based cameras called LISS-I and LISS-II. This decision of payload conguration gave edge to Indian Space Research Organisation (ISRO) as it led to the design and realisation of such multiple high performance sensors with short development time. The payloads were designed to cater to specic requirements of ocean studies, atmosphere, urban and terrain mapping etc. This called for intensive research and development of various technologies. As of now, the sensors team designed and developed 45 payloads catering to Earth Observation (EO) missions (Table-1). Table 1: List of IRS Satellites launched Sr. No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Launch Year 1979 1981 1988 1988 1988 1991 1991 1991 1992 1993 1993 1994 1995 1995 1995 1996 1997 1997 1997 1999 1999 1999 2001 Camera TV TV LISS-1 LISS-2A LISS-2B LISS-1 LISS-2A LISS-2B VHRR VHRR LISS-1 LISS-2 LISS-3 WiFS PAN WiFS LISS-3 WiFS PAN VHRR CCD OCM PAN Satellite BHASKARA BHASKARA IRS-1A IRS-1A IRS-1A IRS-1B IRS-1B IRS-1B INSAT-2A INSAT-2B IRS-1E IRS-P2 IRS-1C IRS-1C IRS-1C IRS-P3 IRS-1D IRS-1D IRS-1D INSAT-2E INSAT-2E OCEANSAT-1 TES Sr. No. 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Launch Year 2002 2003 2003 2003 2003 2003 2003 2005 2005 2007 2008 2008 2008 2009 2010 2011 2011 2011 2011 2011 2013* 2013* Camera VHRR VHRR CCD LISS-3* AWiFS-A AWiFS-B LISS-4 FORE PAN AFT PAN PAN PAN MX-T HySI-T OCM PAN LISS-3* AWiFS-A AWiFS-B LISS-4 LIV HySI IMAGER SOUNDER Satellite KALPANA-1 INSAT-3A INSAT-3A RESOURCESAT-1 RESOURCESAT-1 RESOURCESAT-1 RESOURCESAT-1 CARTOSAT-1 CARTOSAT-1 CARTOSAT-2 CARTOSAT-2A IMS-1 IMS-1 OCEANSAT-2 CARTOSAT-2B RESOURCESAT-2 RESOURCESAT-2 RESOURCESAT-2 RESOURCESAT-2 YOUTHSAT INSAT-3D INSAT-3D

*Expected year of launch

A synopsis of technological evolution of electro-optical payloads in the last 25 years is shown in Figure 1. The evolution resulted in improvements in spatial resolution from 1 km to less than a meter. The design and realization of space borne electro-optical payloads is highly challenging and an involved process. User requirements are the main driving factors for design of electro optical payloads. The varied user requirements are synthesised and translated to sensor requirements. These sensor requirements are analysed vis-a-vis mission constraints such as orbit constraints, satellite resource constraints (attitude, power, volume, thermal, data transmission capabilities,

etc.), available technologies (components, materials, fabrication and test methods etc). All these constraints call for trade off studies, which ultimately lead to optimum sensor conguration and specications. Once sensor system configuration is finalized,
Fig. 1: Evolution of Indian EO sensors

of electro optical payloads poses multitude of challenges including, component availability,

tolerances in the components/ fabrication/ assembly process, complex interplay of subsystems etc. Integration and checkout of the sensors also require considerable efforts in realization of a payload. The Bhaskara-1 TV cameras required development of complex technologies like super Vidicon tubes, mechanical and electronic shutters, high voltage circuitry (14 KV), low noise video processing, discrete components based ADC etc. It was a great challenge to simulate the onboard problem observed after rst switch on and in predicting that the camera operation will be successful if switched on after about a year in the orbit. Great deal of system studies were carried out in IRS-1A arriving at the conguration, selection of push-broom imaging approach, modular conguration for four bands of LISS I & II, selection of state of the art 2000 element linear CCD etc. It was a challenge to build the required condence to use MIL grade components as hi-rel components were not available. For the rst time, TTL logic devices were used to meet speed requirements. Aligning all the four optical and detector assemblies such that they all image the same ground trace simultaneously, was a daunting task accomplished. The interplay of mechanical and thermal stresses onto optical alignment were grappled with, after signicant efforts. Concept of test benches and computer based subsystem and payload checkout and evaluation were introduced. Based on the IRS-1A/1B experiences of using the data, three tier imaging capability was built on IRS-1C platform, with three payloads i.e. PAN (sampler), LISS III (Medium resolution) and WiFS (wide swath). Three mirror off-axis un-obscured reective telescope was designed and developed for the rst time. 4000 and 6000 element CCDs and high speed low noise electronics systems were developed. Realisation of INVAR structure from a forging was adopted to provide dimensional stability. For the rst time, lenses were designed and realised indigenously for WiFS. Shortwave infrared channel was introduced in LISS III. Realisation of 2100 element hybrid SWIR detector with InGaAs photo diodes and silicon ROIC was a great technological feat accomplished successfully. Resourcesat-1 (RS-1) saw upgradation of 3 tier imaging system of IRS-1C/D. PAN camera (5.8 m) is converted to
Fig. 2: Three-tier Imaging System

3 band multispectral camera with innovative

21

N N R M S

B U L L E T I N

realization of payload is initiated. Realization

M A R C H

2 0 1 3

reliability requirements (life, confidence

splitting of eld and introduction of 12000 element CCDs along with band selection lters. LISS III camera had improved SWIR band resolution of 23.5 m compared to 70 m resolution earlier. WiFS was upgraded to AWiFS with 56m resolution compared to 188m and with four bands instead of two bands earlier. Also, SWIR band was incorporated. Efforts in Resourcesat-2 (RS-2) payloads paid off well as radiometric performance of the imageries were improved by implementation of data compression schemes like Differential Pulse Code Modulation (DPCM) and Multi Linear Gain (MLG) without changing S/C mainframe interfaces. RS-2 also saw a signicant miniaturization of electronics hardware (Table 2). Resourcesat-3, also a 3 tier imaging system is being planned with advanced technologies and enhanced performance. This will provide PAN, multi-spectral and hyper spectral data. It will also carry atmospheric correction sensor. The evolution of three tier imaging systems is shown in Tables-3 to 5. Table 2: Miniaturization from Resourcesat-1 to Resourcesat-2 Parameters Volume (in )
3

Resourcesat -1 11296 107 223 131 29

Resourcesat -2 3082 32 180 46 12

Weight (kg) Power (W) PCBs Packages Table 3: Sampler Evolution Parameter/ Sensor name Spectral range() Channels Resolution (m) Swath(km) Quantization IRS-1C, 1D PAN 0.5 0.85 1 5.8 70 6 RS-1 LISS IV 0.52 0.86 1/3 5.8 70/23.5 7

RS-2, RS-2A LISS IV 0.52 0.86 1/3 5.8 70/70 10

RS-3 Enhanced Sampler 0.45 0.9 1(PAN)+4(Mx) 1.25 /2.5 60 10 0.4 2.5 >200 30 60 10

Table 4: Medium Resolution system Evolution Parameter/ Sensor name Spectral range() Channels Resolution (m) Swath(km) Quantization IRS-1C,1D LiSS III 0.52 1.7 4 23.5 (VNIR) 70 (SWIR) 140 7 RS-1 LISS III* 0.52 1.7 4 23.5 (VNIR & SWIR) 140 7 RS-2, RS-2A LISS III* 0.52 1.7 4 23.5 (VNIR & SWIR) 140 10 RS-3 Enhanced LISS III 0.45 1.7 5 10 (VNIR) 20 (SWIR) >200 10

Table 5: Wide Swath System Evolution Parameter/ Sensor name Spectral range() Channels Resolution (m) Swath(km) Quantization IRS-1C,1D WiFS 0.62 0.86 2 188 740 10 RS-1 AWiFS 0.52 1.7 4 56 740 10 RS-2, RS-2A AWiFS 0.52 1.7 4 56 740 12 RS-3 Enhanced AWiFS 0.45 1.7 5 20 >600 12 Atmospheric Corrector 0.4 2.5 >200 240 >600 12

providing two views of the same ground trace with in a pass with 2.5 m planimetric resolution and 5 m vertical resolution. It helped in generating digital elevation model (DEM) for the country. Three mirror telescope was upgraded with 500 mm diameter to ensure adequate signal to noise ratio. High speed ECL logic devices and very high bandwidth current feedback ampliers were used to meet speed requirements. Each of these long focal length cameras weighed 250 kg even after mass optimization. The meter class imagery required for mapping terrain proles and cadastral mapping are met with Technology Experiment Satellite (TES). It was a great step in Indian remote sensing that we had designed and launched such high resolution system in about 2 years with many new technologies. For the rst time, R-C type telescope, mission control rmware (for step and stare operation of camera) and steerable antennas were designed and large size mirror (560 mm), was developed indigenously. This was carried forward in Cartosat-2 series satellites. The payload optics was made very light even with 700 mm dia aperture. Carbon Fibre Reinforced Plastic (CFRP) based camera structure and kinematic mirror mounts were developed to minimize the stresses onto mirror due to assembly stresses and thermal variations. The assembly, integration and testing of this camera was an uphill task. It was required to assemble telescope under gravity conditions for operation under zero gravity conditions. It was a great learning to simulate the optical performance of the telescope using opto-mechanical models to the accuracies of few nano-meters and experimental verication of the same. These technologies are being upgraded to improve the resolutions in Cartosat-2C series and Cartosat-3 series satellites, planned to be launched in near future. CCD technology is being replaced with TDI technology and a large scale miniaturization of electronics and signicant improvements in weight reduction of optics are planned. Table 6 shows evolution of high resolution imaging systems. Table 6: High Resolution Imaging System Evolution Parameter/ Sensor name Spectral range() Channels Resolution (m) Swath(km) Quantization TES PAN 0.5 0.85 1 1 16 7 Cartosat-2, 2A, 2B PAN 0.5 0.85 1 0.8 10 10 Cartosat-2C, 2D, 2E PAN 0.45 0.9 1 0.65 10 11 MX 0.45 0.86 4 2 10 11 PAN 0.45 0.9 1 0.25 16 11 Cartosat-3, 3A, 3B MX 0.45 0.86 4 1 16 11 HySI 0.4 2.5 >200 12 5 11

The design of Ocean Colour Monitor (OCM) payload saw many rst time developments. To cater to 2 day repetivity and narrow band requirements, large eld of view matched Tele-centric lenses were developed for 8 spectral bands. As the sensor was designed to image land and ocean, large dynamic range CCD and 12 bit digitization were developed. It was a challenging accomplishment of modelling and controlling noise to realize true 12 bit consistent performance from bread board level to on-orbit operations. The system performance was limited by only photon noise. OCM was own on Oceansat-2 with small variations in band denitions. Figures 3 and 4 show some of the complexities of OCM sensors in terms of signal and noise. OCM-3 is being designed with enhanced system performance with 13 bands in VNIR. In addition, 2 TIR bands are being incorporated. This requires state of the art developments in all subsystems. Table 7 shows a summary of Oceanographic payloads. 23

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Cartosat-1 was ISROs rst dedicated stereoscopic mission with Fore and Aft cameras

Table 7: Oceanographic Payloads Evolution Parameter/ Sensor name Spectral range() Channels Resolution (m) Swath(Km) Quantization 12 OS-1 OCM-1 0.4 0.9 8 360 1440 12 OS-2 OCM-2 0.4 0.9 8 360 1440 12 OS-3 OCM-3 0.4 0.9 13 360 1440 14 During this time frame, electro-optical payload developments for meteorological applications from GEO platforms also blossomed. Very High Resolution Radiometer (VHRR of INSAT 1/2/3 series and Kalpana) sensors were developed for weather monitoring. In these developments, we have scaled many technological challenges like cryogenically cooled detectors, hi-rel scan mechanism with accurate positional accuracy and long life, very large dynamic range signal processing, synchronous logic etc. In such systems, the back ground signal occupies the
Fig. 3: Typical Radiances Encountered by OCM

TIR 10.5 12.5 2 1080 1440

major part of the output signal and it was challenging to extract useful signal from the detector output. In addition to whisk broom imaging using VHRR, push-broom imaging with linear CCDs was introduced from GEO orbit to provide better information on cloud, snow and other land features. Development of INSAT-3D has been a signicant stride in weather monitoring. Six channel imager and 19 channel sounder are state of the art instruments. Development of GEO Imaging Satellite (GISAT) has been initiated with multi resolution, multi- instrument capability in one satellite to cater to land, ocean and atmosphere applications. This development requires harnessing of a large number of

Fig. 4: Noise power characteristics of OCM

state of the art technologies. Data from this payload is expected to help usher in new application areas. Figure 5 shows evolution of Indian weather monitoring instruments. Table 8 gives a summary of performance of these instruments. Though the space design always demands designers to optimize size, weight and power, there was a concerted effort to design miniature satellites and payloads. Payload team made signicant changes in the approaches of subsystem and system design and realized 4 band camera (equivalent or better than LISS II of IRS-1A in performance) in about 6 kg by adopting innovative designs, state of the art technologies and components. The same was further brought down to <1 kg subsequently. Figure 6 summarises the achievements in miniaturisation of electro-optical payloads.

Spectral range() Channels Resolution (km)

0.5 12.5 2 2.75 (VIS), 11 (IR)

0.5 12.5 2 2 (VIS), 8 (IR)

0.5 12.5 3 2(VIS), 8 (IR)

0.5 1.7 3 1

0.5 12.5 6 1 (VIS), 4-8 (IR)

0.5 15 19 10

0.45 13 12 50 m VIS, 1.5 (IR)

0.4 2.5 >200 <0.5

Hyper spectral imaging is one of the powerful tool for characterizing various species and land/ ocean and atmospheric structures. The sensors team at Space Applications Centre took initiative and developed imaging spectrometers. Spectral dispersions were achieved with plane grating and wedge lters. These instruments were own on aircraft and on satellites namely IMS-1, Chandrayaan-1 and Youthsat. Presently the development is in progress with convex grating and also with
Fig. 5: Weather monitoring Instruments Evolution

interferometric approach. Electro-optical sensors have been developed for observing land, ocean and atmosphere (Figure 7). The continuity sensor systems are already dened as per programmatic requirements. Twenty ve years of glorious journey of IRS payloads not only enabled development of remote sensing applications for societal benets, but also has paved ways for future developments. During this time period, ISRO achieved leapfrogging jump in the satellite based electro-optical technologies with the design and development of state-of-the-art payloads and provided sustainable operational services for earth observation. The experiences of developing earth observation electro-optical sensors enabled realization of hi-rel planetary payloads also. Development of advanced and

Fig. 6: Electro-optical payload miniaturization- an evolution

Fig. 7: Imaging spectrometers- an evolution

complex systems like very low light detection systems, very high resolution systems, laser

based active remote sensors, ne line VIS/IR spectroscopy for atmosphere and ocean monitoring, onboard signal and image processing systems, intelligent and smart autonomous payloads, customized thematic payloads etc., is initiated. These ambitious plans will augment the existing IRS eet to usher in a new era in the eld of remote sensing for earth observations. 25

N N R M S

B U L L E T I N

M A R C H

Parameter/ Sensor name

VHRR

Enhanced VHRR

3-channel VHRR

CCD

Imager

Sounder

GISAT MX HySI

2 0 1 3

Table 8: Weather monitoring Imaging System Evolution

Over the years, a system of procedures and methods have been evolved to deal with the task of electro-optical payload realization all through the stages of system design detailed design analysis realization decision making, onorbit operations, etc. The professional working culture has enabled induction of new personnel into the fold and the new blood helped to go from strength to strength. Another reason behind the exemplary success of all these complex technologically challenging instruments is due to innovation, seless and dedicated efforts of large number of professionals at every level.

Acknowledgement
It was possible to summarise the developments due to excellent documentation, records and discussions among various persons. The author wishes to thank sincerely all the members of electro-optical payload development team at Space Applications Centre. The encouragement received from Director, SAC is gratefully acknowledged.

EvOLUtION AND GROWtH OF SENSORS, OPtICS AND THIN FILMS tECHNOLOGIES At LAbORAtORY FOR ELECtRO-OPtICS SYStEMS (LEOS)
Kamalakar J A Laboratory of Electro-Optic Systems, Bangalore 560 058, India Email: kamalakar@leos.gov.in

Introduction
Laboratory for Electro Optics Systems (LEOS) is the lead centre for the delivery of the state of the art attitude sensors, and cameras for spacecraft control and also the lead centre for delivering advanced reective and refractive optical systems for spacecraft electro optical payloads. The laboratory is also involved in technology developments in payloads for planetary sciences, laser based instruments, MEMS and detector developments. The Indian Remote Sensing Satellites (IRS) own over the years and the new series of remote sensing satellites used many sensors and optical elements designed and developed at LEOS

Sensors?
Sensor responds to a physical stimulus (heat, light, sound, pressure, magnetism, motion) and determines a satellites position or attitude. The sensors are used at different phases of spacecraft control operations and also for rendezvous and docking. Solid state spacecraft sensors encompassing several detector elements play an important role in all spacecraft, manned or unmanned. Sun sensors, earth sensors, magnetometers, star trackers are essential for orbit and attitude control of the spacecraft. The exact, reliable operation of these sensors is a pre-requisite for the success of the mission. This article provides a glimpse of the evolution and technology growth over the last twenty ve years at the LEOS.

Technology Growth in Sensors Area

Indian Space Research Organisation (ISRO) started the remote sensing in late seventies by launching rst experimental satellite Aryabhatta followed by Bhaskara-1 and 2 and later by IRS series of satellites. The sophisticated and successful sensors of today, arose from humble beginnings at ISRO Satellite Centre (ISAC) way back in 1970. This was before the rst Indian Satellite Aryabhata. Aryabhata carried sun sensors, magnetometer, temperature sensors. The successful on-orbit performance of these sensors gave condence to build variety of sun sensors over the years such as twin slit sun sensor, ne analog sun sensor, coarse analog sun sensor, digital sun sensor, precision yaw sensor, solar panel sun sensor. These sensors measure the sun aspect angle and provide essential information for control operations. Sensors group at LEOS mastered and indigenized all the parts required for the sun sensors that is the solar cells, the optics, etc. and to-date own over 500 of them in different satellites.

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

In the beginning of 1977, the sensors team stepped into advanced sensors. The major activity of design of horizon crossing sensor, the rst infrared sensor was taken up by the group. Several challenges were faced in handling the IR detectors, design of its processing electronics and the design of the test facilities. Along the journey towards advanced sensors, the development team came across many failures and learnt many lessons. These development process was a mixed feeling of pleasure as well as frustration. After many design and development efforts, the rst horizon crossing sensor was own in 1979. This worked on
Fig. 1: Sun sensors own so far

14-16 micron Infra Red (IR) band and has the advantage of providing the information night and day. Its successful operation in Bhaskara was a great satisfaction to the entire team. The horizon crossing sensor success led to attempting higher challenges and this culminated in proposing the widest range of sensors for the APPLE Indias rst communication satellite. This satellite carried a variety of sensors (i) Magnetometers (ii) Three types of sun sensors (twinslit, digital sun sensor and solar panel sun sensor) (iii)Three varieties of infrared sensors (the horizon crossing sensor for initial phase, Static horizon sensor, scanning infrared sensor for an orbit phase) Figures 1 & 2. Apple programme gave the maximum challenges to the entire electro optics team and was the stepping stone for the design and development of state of-

Fig. 2: IR sensor and magnetometer

the- art sensors that are being presently made at LEOS. The static horizon sensor and the scanning sensors were totally new designs; horizon sensor was based on absolute radiance sensing using thermopiles and the scanning sensor was based on torsion bar based scanning mechanism and used thermistor bolometers. Designing the absolute radiance static sensor, which had to receive the earth infrared radiation in 14-16 micron band against the strong thermal signal background of the spacecraft and process the signal to produce attitude errors was a difcult task and it was achieved after many development efforts. The critical technologies for the scanning sensor were the 0.57 mm dia torsion bar based scan mechanism, the high resolution incremental optical encoder, the analog and digital processing electronics. These were successfully achieved and todays Indian communication satellites carry three of these infrared earth sensors which are the true eyes of the satellite operating 24 hours a day awlessly.

entire sensor team stationed at Sriharikota when the rst indication of the Apple satellite reaching GEO orbit was given only by horizon crossing sensors when the Apogee Kick Motor (AKM) was red and all other indications were absent for a short period. The successful performance of the Apple sensors gave enough condence to start the development of Indias operational communication satellite INSAT series. To
Fig. 3: Earth sensors

date over 300 sun sensors and over fty earth sensor are own in these satellites (Figure 3). It was satisfying achievement when 50th IR scanning sensor was built and own in W2M satellites. Indias next ambitious programme of the nineteen eighties, the high resolution cartography and remote sensing satellites termed as IRS1 series, which is celebrating its 25 years of operation now, called for high accuracy attitude sensors. Sensor group at LEOS started the star tracker development in the year 1985, which is once again a totally new development with multiple critical technologies. This sensor works on the principle of sensing the low intensity stars and comparing them to an onboard stored star catalog and determining the attitude errors. The multiple refractive element optics design and fabrication, the CCD detector based processing electronics and the digital signal processing using 16 bit processor and the necessary ADA based software were developed after much iterations and highs and lows. Todays all remote sensing satellite carry advanced high accuracy star trackers. An advanced version with a 32 bit processor based electronics was recently own successfully on Youthsat.

Fig. 4: Development of star sensors

Technology Growth in Optics and Thin Films Areas

The rst operational satellite IRS-1A was launched on 17th March, 1988 with two camera systems viz. LISS-I (Linear Image Self Scanning) and LISS-II aimed to meet the specic application 29

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

It was a very proud moment for the

requirements of our country in the elds of agriculture, forestry and geology. The camera systems are operated in four spectral bands for multi spectral imagery with separate optical system for each spectral band. LISS-I and LISS-II camera systems have separate optical systems operating in the same spectral bands with resolution of 73 m and 36.5 m. To meet the requirement of ISROs satellite program, the development of reective and refractive type optical systems started in LEOS way back in 1990. The Applied Optics Team at LEOS, has evolved a comprehensive technology base, comprising of complex optics and opto-mechanical designs, precision optics fabrication, complex coating process technologies for anti-reection coatings, spectral band pass lters and attenuation lters (graded/conventional), optics systems test and evaluation demonstrating the image quality performance as per specications and space qualication of the lens systems against stringent requirements of space applications. The development and successful realization of Wide Field Sensor (WiFS) assemblies for IRS-1C and ID with two spectral bands in visible regions (Band-3 and Band-4) for multi-spectral imaging meeting with diffraction limited performance with excellent image stability have demonstrated the capabilities for future complex multi-element optical systems. The same technology has been extended to Short Wave Infra Red (SWIR) region, designated as Band-5, and successfully own in IRS-P3 mission along with visible bands. The WiFS camera system provides resolution of 188 m and swath coverage of 770 km with 5 days repetivity. The next major milestone in the development of multi-element optical system was the Advanced Wide Field Lens (AWiFS) for Resourcesat-1 (IRS-P6) mission. The AWiFS optical system comprising of four spectral bands (B2, B3, B4 & B5) with individual optics head for each spectral region dened by the interference lter. The enhancement of Effective Focal Length (EFL) from 60 mm to 137.9 mm as that of WiFS leads to the increase in size to maintain the wider eld of view. The stringent requirement of image
Fig. 5: Refractive telescope optics development

format matching (within 2m), image stability (within 1m), and diffraction limited performance and radiometric performance

viz, transmission and veiling glare are met with respect to thermal and vibration environments. Dual head conguration has been adapted to achieve the swath of 740 km with resolution of 56 m. In order to match the resolution of LISS-III visible bands, the development of long focal length (451.7mm) LISS-III SWIR optical system was taken up and realized meeting with performance requirements. All these optical system are own in Resourcesat-1 satellite. For the continuity of the Resourcesat-1 programme, indigenously developed LISS-III (B2, B3, B4 & B5) optical systems in addition to AWiFS has been delivered to Resourcesat-2 mission with performance requirements in all respects. In order to meet the small size and volume requirements for Indian Mini-satellite (IMS) programme, LEOS has developed a compact sized (250 mgm), high performance (70 lp/mm) 4 bands (B1, B2, B3 & B4) multi-spectral (MSC) optical systems including the development of Band-1, covering the wavelength region 450 to 520 nm. These MSC optical system provides 36 m resolution and swath coverage of 151 km. The MSC optical system has been own in

(HYSI) imaging optics meeting with the performance requirements in all respects for IMS-1 and Chandrayaan-1 Mission. The development of AWiFS and LISS-III optical systems is in progress for Resourcesat-2A satellite programme. The development of multi-elements optical system with aspheric surface, long focal length with wide eld of view and short focal length with larger eld of view to meet the ISROs future requirements for natural resource survey and oceanographic applications is under pipeline (Figure 5). The development of reective type optical systems started paralley in the LEOS, and the rst set of mirrors have been realized for VHRR payload of INSAT-2C mission. The mirrors have been nished to the surface accuracy of /50 RMS by manual polishing and testing using conventional knife edge method (Figure 6). With the procurement of advanced infrastructures viz, computer controlled polishing machine, robotic polishing machine and advance interferometer system for characterization, successfully corrected the nished surface to the high accuracy for AVHRR payload. The development of 500 mm solid concave hyperboloid primary mirror and secondary convex hyperboloid mirror and eld corrector optics within the short span of 6 months has demonstrated the capability of realizing high resolution large optical system within the country. Establishing the test model philosophy with indigenously developed zero gravity test bed for removing gravity related errors, individual testing of eld corrector optics with auxillary optical and integrated telescope system test are all commendable efforts of LEOS. The telescope optics performance has been demonstrated for all eld points and was successfully own in TES mission thus providing 1 m resolution imagery. The fabrication and testing of light weighed (80%) primary mirror for VHRR imager and sounder payload system for INSAT-3D mission was demonstrated for the rst time. The same process technology was extended to fabricate the large size light weighted (705 mm diameter) primary concave hyperboloid mirror and 200 mm diameter convex hyperboloid secondary mirror for Cartosat-2 series missions with 0.8 meter resolution. The model philosophy for testing the aspheric surfaces using null method and removing gravity effect using large size zero-g test bed have been established. The performance of the telescope with coatings developed at LEOS was demonstrated after integration and was successfully own in Cartosat-2 and 2A missions. More details on system specifications of LEOS elements are given in Table 1. The realization of 380 mm solid primary and secondary aspheric mirrors with metal holders for UVIT payload for Astrosat mission was a great technology challenge due to its operations in ultra violet regions which was successfully completed. Two sets of telescope mirrors (Far UV and VIS channel) have been realized meeting with performance requirements in term of high surface accuracy
Fig. 6: Reective optics development

31

and surface roughness and delivered.

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

IMS-1 mission. LEOS has developed a high performance and compact sized (260 gm) Hyper-spectral

Table 1: Systems Specications of LEOS Elements Sensor Specications Specication Oscillatory Earth Sensor 250 0.03 Digital -200C +500C Static Earth Sensor 12.50 0.03 Analog -200C +550C Conical Scanning Sensor 450 0.05 Analog / Digital -100C +500C Dual Cone Scanning Sensor 200 0.05 Digital -100C +550C Digital Sun Sensor 320 0.1 Digital -400C +500C Coarse Analog Sun Sensor 400 0.5 Analog -200C +550C
62x60x42

Fine Sun Sensor

Four Pi Sun Sensor 4 5 Analog -300C +800C

50x50x50

Star Sensor MARK-II 200 x 200 0.0028 Digital -00C +400C 219 x 358 3.0

Range Accuracy (0) Output Operating Temp.

600 0.5 Analog -100C +600C

60x60x30

Size (mm) 250x150x120 150x150x170 110x165x110 130x230x180 150x170x70 Weight (kg) 2.5 3.0 2.0 3.2 0.85

0.035

0.08

0.075

Refractive Systems Specication Field of View EFL F No. Resolution Format Matching Collinearity Stability VG Distortion No. of Bands WiFS 13
0

LISS III 5
0

AWiFS Dual Head 25

0

Star Sensor 14.50 48.5 1.5 60 12m Blur dia -<0.003 mm <2% <0.01% 1

56.47 mm 6.0 188 m <0.001 mm <0.001 mm <2% <0.05% 3

347.5 mm 4.5 23 m <0.001 mm <0.001 mm <2% <0.05% 4 Reective Systems

137.5 4.5 <56 m <0.001 mm <0.001 mm <2% <0.05% 4

PAN LISS-IV 2.50 980 mm 4.5 <5 m ----PAN Band

Advanced PAN 0.80 1904 mm 3 <1 m ----PAN Band

VHRR 223.4rad 1594 mm 7.85 8km (IR) 2 km (vis.) ----3

VHRR Sounder --310 mm aperature <1 km ----19

VHRR Imager --310 mm aperature <1 km ----6

Cartosat Series 0.50 5600 mm 705 mm aperature <0.8 mm ----PAN

The development of complex optical systems requiring large size 1.2 meter, light weighted aspheric primary mirror for Cartosat-3 missons for 0.25 m resolution imagery and complex eld corrector optics for VIS and TIR band for GISAT program is in advanced stages of development.

optics and thin lms activities were started as a support to the sensors development. By 1992, sufcient infrastructure and expertise were developed in the visible and IR optics areas and the major designs of thin lms for the optical payloads of TES satellite were taken up and successfully delivered. Today the state of the art, design, fabrication and test facilities for thin lms at LEOS are supporting a variety coatings of refractive and reective optical elements to all the payloads built in ISRO. coatings on MLI were successfully developed. In addition to the above, efforts in the development of laser based sensors, ber optics gyros, etc., are the source of pride to the sensors group at LEOS. It is the only place in the world where variety of state- of -the -art sensors i.e. sun sensors, star sensors, infrared sensors, laser sensors, bre optics gyros, magnetometers are all done under one roof, providing service to all the spacecrafts own. In the recently own Chandrayaan mission LEOS has contributed the scientic pay load Lunar Laser Ranging Instrument (LLRI) which mapped the terrain of the moon and generated 3D imageries of the craters of the moon. LEOS is working on the new scientic payload Laser Induced Based Spectroscopy (LIBS), to be own on Chandrayaan2, to study the minerals on moon. It is planned to use MEMS based inclinometer, developed in house for using lunar lander / rover and is also developing Lyman Alpha Photometer (LAP), to be own in Mars Mission for the study of the escape process of MARS. The motto of the teams at LEOS is to be ever vigilant and provide the best and reliable optical systems since they are entrusted with the most critical responsibility of the satellite, that is the delivery of optics and sensors, which are the eyes of the satellite. In addition all anti-reection coatings, lters and lter coatings, protective coatings, ITO coatings, germanium

33

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Meanwhile the thin lms activities at LEOS also were progressing at a good pace. Initially the

EVOLUTION OF MICROWAVE REMOTE SENSORS IN ISRO

Tapan Misra Space Applications Centre (ISRO), Ahmedabad - 380 115, India Email: misratapan@sac.isro.gov.in

Introduction
Over the last almost three decades, Microwave Remote Sensing (MRS) programme has evolved in Space Applications Centre (SAC), ISRO (Figure 1). The beginning of MRS programme was made way back in early part of 1980s, in concurrence with optical remote sensing programme. Though, all weather and day-night operation capability of microwave remote sensors, was understood at that time, its fructication needed longer incubation period precisely because of the requirement of development of an array of complex technologies and algorithms. Development of this programme called for higher level of investment in terms of both nancial and technological implications. Over the period, MRS programme evolved into two basic paths of applications, namely atmospheric and oceanographic applications and imaging applications.

Atmospheric and Oceanographic Sensors

Both passive and active microwave remote sensing instruments have been/are being developed to meet the requirements of atmospheric and oceanographic applications. From the modest beginning of development of Satellite Microwave Radiometer (SAMIR) payloads for Bhaskara series of satellites for scientic understanding, payloads like Multi Frequency Scanning Microwave Radio Meter (MSMR) and Scatterometers are designed for providing operational services. At the same time, ISRO initiated collaborative development of payloads, like Meghatropiques, with established space agencies, to consolidate expertise available in the development of complex technologies. SAMIR Payload Satellite Microwave Radiometer (SAMIR) ushered microwave remote sensing technology in ISRO. It was own on Bhaskara-1 and -2 satellites in 1978 and 1981. SAMIR in Bhaskara-1 had two channels in 19.35 and 22.35 GHz. SAMIR payload in Bhaskara-2 had additional channel in 31 GHz. SAMIR provided the rst glimpse of global water vapour and liquid water content with indigenous technology. Bhaskara was a spin stabilized satellite and this fact was effectively used to congure this vertically scanning radiometer. This radiometer was congured in Dicke conguration in which scan view of the open sky was used as cold calibration input. It provided a footprint of 150-250 km with sensitivity of 1-2K. Bhaskara satellites had two spinning modes namely around roll axis or pitch axis. In the rst mode of spin, SAMIR provided a swath of 1500 km from 777 km altitude. In second mode of

of measurement at Nadir. Figure 2 presents photographs of Bhaskara-1 and -2 satellites and a sample brightness temperature product from Bhaskara-1. MSMR Onboard Ocansat-I

payload along with Ocean Colour Monitor (OCM) onboard Oceansat-1 (IRS-P4) satellite, I S R O s f i r s t s a t e l l i t e d e d i c a t e d f o r oceanographic applications. The satellite was launched on May 26, 1999 from Sriharikota Range (SHAR) of Indian Space Research Organisation (ISRO). MSMR was a four-frequency, dual-polarised scanning microwave radiometer to measure brightness temperatures of the earth surface. This was used for deriving parameters like sea surface temperature, wind speed, water vapour content, etc. from the measured brightness temperatures. The system was designed for four frequencies viz., 6.6, 10.65, 18.0 & 21.0 GHz in order to cater to the above applications. The mission specications of MSMR is shown in Table-1. Figure 3 shows integrated view of MSMR payload. MSMR was congured as typical Dickey radiometer, with fixed, multifrequency feed and reector scanning at
Fig. 1: Evolution of ISROs microwave remote sensing programme

the rate of 11.173 RPM, providing collocated measurement at an incidence angle of 49.8o. For the purpose of onboard calibration, two sky horns were used as cold source. One of them was for 6.6 and 10.65 GHz frequencies and the other for 18.0 & 21.0 GHz. Constant integration time of 18 ms was provided to

Fig. 2: Photographs of Bhaskara-1 2 and sample output from SAMIR payload

all the channels and they were sampled uniformly at an interval of 9 ms.

35

N N R M S

B U L L E T I N

MSMR was the co-passenger

M A R C H

2 0 1 3

spin, in every scan, it provided single sample

Table 1: MSMR mission specications Spacecraft Orbit Altitude Swath Repetivity Frequencies Polarisation Grid size Sampling Data rate Payload Weight Payload Power IRS-P4 Sun Synchronous 720 km ~ 1360 km 2 days 6.6, 10.65, 18.0 & 21.0 GHz V & H for all frequencies 150 Km (6.6 GHz) 75 km (10.65 GHz) 50 km (18 and 21 GHz) 12 bits 5.6 kbps 130 kg 80 Watts MSMR provided routinely 48 hour global data product, in terms of brightness temperature, Sea Surface Temperature (SST), wind speed, water vapour content and cloud Liquid Water, on operational basis from National Remote Sensing Centre (NRSC), Hyderabad, India. A typical glimpse of the global product is shown in Figure 4. The payload data was used for various experimental usage like soil moisture, Antarctic ice study etc. One such study result on variation of Ice cover Antarctica is shown in Figure 5.
Fig. 3: Integrated MSMR assembly in clean room environment

Ku-band Scatterometer on-board Oceansat-2 A Ku-band Scatterometer was built in ISRO to gather the information about the near surface winds over oceans at a global level. This payload was own in Oceansat-2 satellite along with OCM (launched on 23rd September, 2009). The wind eld measurements from scatterometer form a very important input to the global weather forecasting system. Table 2 gives the brief mission specifications for the

Fig. 4: Glimpse of routine products available from MSMR data

scatterometer payload.

100 W peak and 1.35 msec duration through a dual polarized offset reflector antenna, spinning at the rate of 20.5 RPM around yaw axis. The Vertical and Horizontal (V&H) beams are pointed at off-nadir pointing of 49.4o and 42.6o respectively (Figure 6). All the regions

azimuthally divergent four measurements of radar cross section of capillary waves riding on the gravity waves of the sea. Within the inner swath of approximately 1400 km, wind direction can be measured umambiguously. However, in the outer swath, because of availability of only two measurements of radar cross section, wind direction cannot be resolved unambiguously. With the help of auxiliary information like model winds, the directional ambiguity will be resolved in the
Fig. 5: Seasonal Variation of Ice Cover Over Antarctica, Derived from MSMR

outer beam.

Table 2: Major specications of Ku-band scatterometer on-board Oceansat-2 Satellite Altitude Inclination Orbit Frequency Polarization Swath Wind Speed Wind Direction Wind Vector Cell (grid) Size Payload Wight Payload Power 720 km 980 Sun Synchronous 13.515625 GHz HH for inner and VV for Outer beams 1400 km (qualied) 4-24 ms/s 00 to 3600 50 km x 50 km Grid 120 kg 230 watts

Scatterometer operates at 200 Hz, alternate pulses are switched between V and H ports of the antenna. There is a provision of extensive on-board calibration of the system to provide improved accuracy in wind vector measurements. Scatterometer has been launched on 23rd September, 2009. It is already providing global coverage of wind vector at 50 km x 50 km grid every two days and the product is available at every 12 hours interval. 37

N N R M S

B U L L E T I N

within the inner beam coverage provides

M A R C H

2 0 1 3

The scatterometer transmits a pulse of

Artistic impression of the scatterometer payload on-board Oceansat-2 is shown in Figure 7. The payload and satellite is shown in Figures 8 and 9. Global coverage of sigma naught is shown in Figure 10. Representative wind vector product is depicted in Figure 11. This payload has evoked considerable Global interest. Global data dissemination and assimilation in weather forecasting are being provided by NASANOAA and EUMETSAT, apart from NRSC. Meghatropiques In order to provide boost to climate research with high repetivity data over the globe, Meghatropiques satellite was planned
Fig. 6: Viewing geometry of scatterometer

as joint Indo-French collaboration between their premier space organizations : ISRO and CNES. It had three payloads: a. MADRAS (Microwave Analysis And Detection Of Rain And Atmospheric Structures)- Imaging microwave radiometer 18.7, 23.8, 36.5, 89 to 157 GHz

(5 Frequencies and 9 channels)

Fig. 7: Artistic view of the Oceansat-II satellite with scatterometer and OCM payloads mounted on the top deck

b. SAPHIR (Sondeur Atmospherique du Prol dHumidite Intertropicale par Radiometrie) -Sounder for Atmospheric Profiling of Humidity in the Intertropics by Radiometry (6 Channels Around 183 GHz) c. SCARAB (Scanner for Radiation Budget)B ro a d b a n d r a d i a t i o n m e a s u re m e n t instrument in shortwave (< 4 mm) and longwave (> 4 mm) MADRAS is configured as total power radiometer with 65 cm dia antenna and provides resolution from 67 km to 10 km over the frequency bands. Provide measurements over a swath of 1700 km at

Fig. 8: Scatterometer payload in SAC clean room

incidence angles of 55o (Figure 12).

by IRS bus and the satellite was launched on 10th October, 2011, in an inclined orbit of 20o at 867 km altitude to provide repetitive coverage of equatorial region between 30o N and 30o S at every 6 hrs. The artistic view of the payload module of Megha-

in Figure 13 and Figure 14. Figure 15 shows a photograph of Meghatropiques payload,

Fig. 9: Oceansat-2 satellite in SDSC clean room

being prepared for thermovac tests. After more than one month of aw less operation of MADRAS payload, it encountered data strobe errors which resulted in misordering of data transmitted. After 10 months of effort, an algorithm was developed to reorder the data and correct the images. One such example is shown in Figure 16. Data has been released to users from 31st August 2012.

Fig. 10: Global coverage of O obtained from OS-2 scatterometer

M i l l i m e t e r- Wa v e Te m p e r a t u r e Sounding Unit (TSU) A new venture in the series of ISROs polar-orbiting earth-observation missions will be the millimeter-wave atmospheric temperature sounder (Table-3). It enables 2D and possibly 3D temperature proling of the earths atmosphere even in the presence of clouds and hydro-meteors using the absorption spectrum of oxygen (which is uniformly distributed in the earths atmosphere) in the millimeter-wave regime. The theory of radiative transfer relates the passive measurements of radiance at the top of the atmosphere with the emission or equivalently the physical temperature of

Fig. 11: Sample global wind vector product from OS-2 scatterometer

each atmospheric layer. The relation is called a weighting function (Figure 17a). Therefore, using a nite number of channels, vertical 39

N N R M S

B U L L E T I N

Tropiques Payload and Satellite is shown

M A R C H

2 0 1 3

All the three payloads are carried

temperature prole can be retrieved with a trade-off of vertical resolution, altitude and accuracy. Such proles serve as seed for studies in boundary-layer interactions, atmospheric motion vectors, the diurnal cycle and so forth. The radiance measurements are also routinely fed into numerical weather prediction models and used for global climate monitoring and modelling purposes. A novel set of channels (distinct
Fig. 12: Scanning geometry of meghatropiques

from those of AMSU-A, ATMS and SSMIS) was selected for the TSU. These channels are primarily optimal off-resonance frequencies in the 50-60 GHz oxygen line-absorption spectrum which impose less stringency on bandwidth and sensitivity and yet have adequate overlap of their weighting functions which can be harnessed to achieve the desired vertical resolution. The consortium consists of 11 single-pass-band line-wing channels (Figure 17b) with near-uniform bandwidth (300 MHz) and brightness temperature sensitivity (< 0.5K) and can sound up to 30 km with vertical resolution ranging from less

Fig. 13: Integrated View of Meghatropiques Payloads

than 2 to 6 km. There are 2 double-passband channels near line-centres which can sound up to 40 km with poorer resolution. In addition, there are a couple of radiometric channels at 23.8 GHz and 31.5 GHz for estimation of surface-contamination as well as for sensing of water vapor and rain. It has been proposed to y the TSU onboard Oceansat-3 as a co-passenger of a Ku-band scatterometer. A Schematic view of Scatsat and TSU are give in Figures 18a & 18b. A RF-shield has been designed to protect the radiometer (on the earth-view panel) from possible interference with the radiating

Fig. 14: Artistic View of Meghatropiques on IRS Bus

scatterometer. The satellite will be placed

coverage at a nominal altitude of 725 km with descending and ascending node local times at 12 noon and 12 mid-night respectively and a revisit time of approximately 48 hours. A view of DVM of TSU, under test, is shown in Figure18c. The TSU will be a cross-track scanning sensor in order to achieve good spatial resolution at nadir (40 km x40 km which can be improved to 10 km x10 km by virtue of over sampling) and good calibration accuracy. In a vertical scan-cycle the sensor
Fig. 15: MT payloads being prepared for thermovac test

will look at the earth through an angular window of 450 dening a swath of 1550 km across the sub-satellite track (Figure 19). It will also look towards two targets with a wide temperature difference between them for the purpose of two-point calibration: a blackbody (developed and qualied indigenously) mounted on the satellite that will serve as hot-target and the deep space that will serve as the cold-target.

Fig. 16: 36.5 GHz channel data from MADRAS before and after correction

The TSU has been configured as a pair of total power radiometers (each sensing a polarization H/V) that will operate in

tandem. The redundancy in payload electronics has been implemented by simultaneous measurements of two polarizations of all channels, thereby eliminating the possibility of front-end losses (by switching or division of power) and unfurling the scope of a gamut of studies on atmospheric scatterers as well as surface contamination in low-altitude sounding. Table 3: Basic Sensor Specication of Proposed TSU Sr No 1 2 3 4 5 6 7 Parameter Frequencies (GHz) Polarization Dynamic Temperature Range Look Angle Coverage Incidence Angle (max) Scan Rate (rpm) Swath (Km) Specication 23.8, 31.5,(Surface correction channels) & 50-60 (15 Temp V/H 40 - 3100 K -45.00 to +45.00 520 12.85 (Temperature Channels) 1550 (approx.) 41

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

in a polar sun-synchronous orbit for global

Separate antennas for 23.8/31.5 GHz and the 50-60 GHz band have been the nal choice for a scanning system to achieve uniform beam-width and yet not compromise the beam-efficiency. The dual-reflector Gregorian antenna has been adopted to achieve ~95% beam efciency, and better than -23dB cross-polarization performance within reasonably compact dimensions. Each antenna system consists of a scanning main reector, a stationary secondary reector, a prole-corrugated feed-horn and an othomode transducer for polarization separation. The scanning reflectors of both antenna systems will be placed back-to-back so that their rotation can be synchronised by a common scan-mechanism.
Fig. 17: (a) Weighting functions of TSU channel (b) TSU channel centre frequencies amidst the 50-60 GHz oxyen absorption spectrum

The atmosphere-leaving radiance captured and polarization-separated by the antenna will be fed to a wideband (50-60 GHz) low-noise amplier and down-converted at dc to 5.5 GHz using a sub-harmonic mixer at the front-end. At the backend, the IF energy undergoes power-division using 1:10 and 1:5 Wilkinson power dividers, separation of channels using 17 precise band-pass lters, detection using backward tunnel diode and digitization using 14 bit data acquisition system. An automatic gain control loop is in place to adjust the power level at the input to the detector so that the diode can be made to operate in linear power-to-voltage conversion regime and the dynamic range of the sensor is captured within the ENOB. An array of platinum resistance thermometers will be deployed for sensing the physical temperature of hot-load as well as reectors with a precision of 0.1K. A

Fig. 18: (a) A schematic view of TSU aboard Scatsat- (b) A schematic view of TSU payload conguration

payload controller will generate all timing and control signals for the sub-systems, interface

electronics, format the data grabbed over a complete scan and transfer the same to BDH.

Radar Based Imaging Sensors

ISRO mastered space based imaging
Fig. 18: (c) DVM of TSU under test

capability two decades ago with the launch of IRS-1A on 17th March, 1988. Progressively, optical camera based imaging capability is in the path of continuous improvement both in terms of resolution and swath coverage till this date. In a programmatic approach, the scope of ISROs remote sensing capability was extended to microwave bands after attaining successful grip over imaging technology in optical bands. Synthetic Aperture Radar (SAR) was logical choice for Imaging Radars in microwave bands, considering its all weather and day-

Fig. 19: A schematic view of TSU cross-track scan conguration

night capability. Further, radar response to geometric shape and distribution of features, was realized to be complementary and supplementary to spectral reectance based feature characterization available in optical imaging process. However, development of SAR needed mastering of both sophisticated radar technology and complex SAR processing algorithm. Since space borne SAR was highly cost intensive, airborne version of SAR was embarked upon as cost effective route of mastering SAR technology in all its nuances. Over the period, both airborne and space borne versions of SAR technology development have taken vibrant shape in ISRO

Fig. 20: SLAR system mounted in dakota and a sample of imagery obtained by SLAR

towards harnessing their utilization potential.

Airborne Imaging Radar Development X-Band Side Looking Airborne Radar Imaging Radar development was initiated modestly with development of Side Looking Airborne Radar (SLAR) in X-band (Table-4). This SLAR was basically used for development 43

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

with the spacecraft BMU and the scan control

of radar technology and it was in ying condition from 1985-1989. The view of the SLAR system and a typical scene imaged by this system, is presented in Figure 20. SLAR provided imaging over 2-5 km swath with 50 m -100 m resolution. Compared to todays technology, its capability was primitive; but it acted as a test bed to test various new technologies like the radar itself, data recording system and on-board Quick Look Display (QLD) system. Further, the experience gained by experimenting with this system, acted as foundation for subsequent development of complex SAR technology. Airborne Synthetic Aperture Radar (ASAR) ASAR (Airborne SAR) was the rst proud foray of ISRO in SAR development . The aim of this development was not only technology attainment but also demonstration of application potential. Consequently this system was congured in C-band providing 6 m resolution over 25 km swath. Detailed specications of the ASAR system is presented in Table-4. The ASAR system was used for experimentation since 1997 and for the rst time, a modest narrow swath (2.5 km) Real Time Processor was demonstrated in this system. The high point of this project was development of a sophisticated SAR processor involving motion sensing and compensation. In fact it had a novel two track motion compensation processor which added to the robustness of the system. This algorithm was implemented on a parallel processing system consisting of 8 Xeon processors. A sample image of generated by ASAR along with photograph of the system can be seen in Figure 21. Table 4: Summary specications of airborne imaging radars developed in ISRO SLAR (X Band) Frequency Polarisation Pulsewidth/ Chirp Bandwidth Resolution/Swath 9.6 GHz HH 80/300 ns ASAR (CBand) 5.3 GHz VV/HH Beechcraft (8 km) 20 sec 25 MHz 6-8 m/ 25 km DMSAR (C Band) 5.35 GHz VV/HH Beechcraft (8 km) 20 sec 225,75,37.5, 18.75 MHz 1 m/6 km 3 m/ 25 km 5 m/50 km 10 m/70 km 452 hz 8 kW
2

DMSAR (X Band) 9.65 GHz HH Beechcraft (8 km) 20 sec 450,225,75 MHz 0.5 m/4.5 km 1 m/ 7.7 km 3 m/7.7 km 360-720 Hz 6 kW
2

Platform/ Altitude Dakota (2-3 km)

50-100 m/ 5 km

PRF Peak Power Antenna Length/ Pattern

850 Hz 25 kW

425-525 Hz 2 kW

2 m/ Cosine Weighted 1.3 m/ Cosec

1.3 m/ Cosec

0.8 m/ Cosec2

ASAR has been used to demonstrate ood mapping and monitoring operation in 2003-2004 (Figure 22). ASAR experience provided the impetus for development of an improved version of C-band airborne SAR, DMSAR (SAR for Disaster Management), exclusively for ood monitoring activity. ASAR experience has demonstrated that for quick reaction to ood related emergency, airborne SAR system is more effective than its space borne counterpart. Space SARs, with their xed orbital conguration, cannot ensure quick reaction to ood situation. DMSAR Airborne SAR for Disaster Management DMSAR in C-band was developed with operational requirement in view. To aid ood monitoring and assessment of damage to infrastructure, four resolution modes were provided (Table-4). DMSAR was test own for the rst time

and its effectiveness in ood assessment was evaluated in 2006. From the year 2007, it has been pressed into trial operational service. To support its high resolution and wide swath capability, the ASAR processing algorithm was signicantly modied. A new algorithm called Track Steering algorithm was developed (under process of patenting) to process in high resolution modes. The DMSAR processor is now available on three platforms: a. A parallel machine with 32/64 Itanium
Fig. 21: ASAR system mounted in beechcraft and a sample of imagery obtained by ASAR

processors b. In house developed Near Real Time Processor (NRTP) built around 96 TS1 0 1 p ro c e s s o r s . T h i s i s a p o r t a b l e system, which can be carried in the Beechcraft itself and taken to base camp for processing and dissemination of DMSAR data. c. GARUDA computing grid provided by CDAC. Using this grid, DMSAR data can be processed simultaneously on an array of computing nodes available in academic and research institutions across India. Figure 23 shows the photograph of DMSAR in operation and also a sample of

Fig. 22: Preood and during ood image over Maligaon, Assam, imaged by ASAR in 2004

high resolution imagery obtained by it. Figure 24 depicts a typical ood region, mapped by

DMSAR, over Dharbhanga, Bihar in 2007. Figure 25 shows the devastating breach of Koshi river in Napal in Sept. 2008 as captured by DMSAR. This breach resulted in change of Koshi river track and resulted in devastating ood of Bihar plains. Presently DMSAR in X-band is under test. While C-band DMSAR operates in high incidence angle of 65 - 85 for wide swath coverage, X-band DMSAR is being designed for high resolution operation at low incidence angle of 25 - 55 . RISAT-1 Radar Imaging Satellite, ISROs First SAR Satellite Radar Imaging Satellite-1 (RISAT-1), carrying Indias rst indigenously developed space-borne, C-Band Synthetic Aperture Radar (SAR) payload was launched on 26th April, 2012 by PSLV-C19 ight. 45

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

in Nov. 2005. It was calibrated completely

RISAT-1, after its positioning at 536 km sunsynchronous dawn-dusk circular orbit, was operated on May 1, 2012 and the SAR images of good quality, have been received. Success of mission marks entry of ISRO into a select group of space agencies operating C band SAR. At present only Canadian RADARSAT -2 mission is providing
Fig. 23: DMSAR system mounted in beechcraft and a sample of high resolution imagery obtained by DMSAR

images in this frequency band. However, a number of agencies have been providing SAR images in X band (TerraSAR-X from Germany, Cosmos Skymed from Italy and RISAT-2 from India). RISAT-1 Mission and SAR Payload Configurations RISAT-1 mission is designed to provide SAR images covering the country with a repetivity period of 25 days, at 6 am - 6 pm equator crossing time. It is possible to operate RISAT-1 for up to 10 minutes duration

Fig. 24: Extent of ood from preood image obtained on 23/06/07 and ood image obtained on 03/08/07 over Dharbhanga, Bihar Blue Colour: ooded region Black Colour: perennialy water logged region

in each of its 14 orbits per day around the earth. It is possible to send the image data collected by the payload to ground station in real time or use the onboard solid state recorder and down link later to the ground station. This recorder capability enables imaging of any portion of the globe. The spacecraft design features a very high transmission data rate of 640 Mbits/sec in X-band. This is achieved using frequency reuse of X band with right hand circular and left hand circular polarisations. RISAT payload is an active remote sensing system as it carries its own source of illumination. The payload transmits a series of electromagnetic pulses of radiation in C

Fig. 25: Koshi river breach imaged by DMSAR in 2008

band using an active antenna array of 576

strikes the earth surface and the back scattered signal is received by the receive modules mounted on the antenna and by time correlated processing of this signal, information about the earth surface is deciphered. The large number of transmit receive modules on the antenna panel enables generation of a beam of electro-magnetic pulses in C band. By controlling the phase and number of modules energized, it is possible to change the beam direction as well as beam-width. This is known as electronic beam steering. Beam steering capability enables operation of the payload in a mode called

The system is also designed to send and receive the signals in different linear and circular polarisations. These capabilities make RISAT 1 payload a unique SAR, currently in operation from space. This payload is unique because of its ability to provide a large swath( ~200 kms) images in multi polarization modes. These modes cover transmit in Horizontal polarisation and receive in horizontal polarisation (HH mode). Similarly one can transmit in Horizontal and receive both in Horizontal and vertical polarizations. (HH+HV modes). In similar fashion one can obtain VV or VV+VH. One can even obtain Quad polarization data (VV+HH+HV+VH) by transmitting the signal in H and V in periodic bursts and receiving the data in both H and V polarization simultaneously. RISAT SAR has unique Hybrid polarimetry mode, where signal is transmitted in circular polarization and signal is simultaneously received in H and V
Fig. 26: RISAT spacecraft conguration

polarization.

Figure 26 illustrates overall satellite conguration and Table-5 gives the salient features. RISAT-1 has imaging capabilities in Strip-map and ScanSAR modes with resolution from 1-50 m and swath coverage from 25 km to 223 km, with multi-polarization capabilities. Imaging mode congurations are illustrated in Figure 27 and salient imaging specications are shown in Table-6. The basic imaging modes, identied for RISAT-1 SAR payload, are as follows: Fine Resolution Strip-map Mode-1 (FRS-1): It is based on conventional mode of SAR Strip-map imaging. The orientation of the antenna beam is fixed with respect to ight path so that constant swath (25 km) is illuminated along the ight direction. The intended resolution is 3 m for FRS-1 mode. Coarse Resolution ScanSAR Mode (CRS): This mode allows for a multifold increase of the swath. This is achieved by periodically stepping the antenna beam to the neighboring sub-swaths (in the range direction using 12 beams). Swath in CRS mode would be 223 km with a resolution of 50 m. 47

N N R M S

B U L L E T I N

ScanSAR mode.

M A R C H

2 0 1 3

transmit- receive modules mounted on panel of ~6 meters by ~2 meters. The electromagnetic pulses

Table 5: Salient RISAT-1 spacecraft parameters Orbit Orbit Altitude Orbit Inclination Orbit Period No. of orbits per day Equator Crossing Spacecraft Height Mass Power Max power handling capacity Data rate SSR TT&C Payload down link Power Pointing accuracy Drift rate Attitude knowledge Circular Polar Sun Synchronous 536 km 97.552O 95.49 minutes 14 6.00 AM/6.00 PM 3.85 m 1858 kg Solar Array Generating 2200W and one 70 AH Ni-H2 Battery 4.3 KW 2x160 Mbps (Total 640 Mbps in 2 chains) 240 Gbits (EOL) S-band X-band 70V bus /42V bus 0.05O 5.0 x 10-5 O/sec 0.02 O Medium Resolution ScanSAR Mode

(MRS): This is a 6-beam scanSAR mode, similar to the CRS mode, providing a resolution of 25 m over a swath of 115 km. Fine Resolution Strip-map Mode-2 (FRS-2): This mode has quad polarization capability. Conceptually, this mode is a hybrid of strip-map and scanSAR, where polarization combinations are switched periodically instead of beams. High Resolution Spotlight Mode (HRS): In the spotlight mode, the antenna beam
Fig. 27: Illustration of imaging Modes of RISAT-1

is oriented continuously to illuminate a particular 10 km x 10 km spot on the ground

which can be imaged with 1 m resolution. The spot size can be increased up to 10 km x 100 km. Circular Polarimetric Modes (C-HRS, C-FRS-1, C-FRS-2, CMRS, C-CRS): All the modes mentioned above can be operated in hybrid-circular polarimetric conguration. RISAT-1 has the capability to cover both sides of the sub-satellite track by roll-tilting the satellite. The antenna is mounted on satellite with antenna normal coinciding with satellite yaw axis. Before start of the imaging, the satellite will be roll-tilted by 36 to enable left/right imaging. With this position of the satellite, electronic beam steering will be used to cover the ground range distance of 107 km to 659 km off-nadir covering an incidence angle of 12 to 55.

Swath Coverage Incidence Angle Coverage Image Quality Mode Single Pol HH/HV/VV/VH

Selectable within 107 659 km off-nadir distance on either side 120 550 Polarisation Dual Pol HH+HV/VV+VH Circular (Hybrid) Polarimetry TX: CP Rx: V and H (Experimental) Quad Pol HH+HV+VV+VH

HRS FRS-1

1 m (Azimuth) x 0.67 m (Range) resolution, 10 x 10 Km (10 x100 km Experimental) Spot Min so= -16 dB 3 m (Azimuth) x 2 m (Range)resolution 25 km swath Min so= -17 dB 3 m (Azimuth) x 4 m (Range) resolution 25 km swath Min so= -20 dB 21-23 m (Azimuth) x 8 m (Range) resolution, 115 km swath Min so= -17 dB 41-55 m (Azimuth) x 8 m (Range) resolution 223 km swath Min so= -17 dB 9 m (Azimuth) x 4 m (Range) resolution 25 km swath so= -19dB

MRS

CRS

To provide near-constant swath, elevation beam width is varied from 2.48 to 1.67 by varying the electrical width of the antenna by switching-off certain rows of TR-Modules, when imaging closer to the satellite track and switching-on all the rows when imaging farther off-nadir regions. Total 122 antenna-beams (61 for left- and 61 for right-imaging) have been dened to cover the entire given range. Out of these, any 12 beams can be used to dene a particular imaging session. During imaging, satellite will be steered in both yaw and pitch to align the beam to zero Doppler line to ease SAR processing by reducing Range Cell Migration (RCM) to minimum. The earth viewing part of the antenna is a printed micro-strip patch array. 24 dual polarised linear arrays are printed in each of 12 tiles. Each of these linear arrays, consisting of 20 patch arrays, is fed by one pair of TR modules, each of which is dedicated for one polarisation. The active-antenna conguration will also help tolerate random failure of up to 10% TR-Modules, with only graceful degradation of antenna pattern. Four Tiles make one panel. RISAT-1 antenna has total 3 panels, of which one is xed and adjacent two are deployable. The radiating surface of the antenna is covered with glass wool blanket to prevent heating by earths radiation. Both of the TR pair receives DC power from a miniaturised EPC called Power Control and Processing Unit (PCPU). Each such TR pair is controlled for synchronous operation by an ASIC based TR Control
Fig. 28: Photographs of a representative tile and panel

49

Computer (TRC). All the 24 TR module pairs

N N R M S

FRS-2

B U L L E T I N

M A R C H

2 0 1 3

Table 6: RISAT-1 image quality parameters

in a Tile are managed by a Tile Control Unit (TCU). TR modules, each with maximum pulsed power of 10 watts, along with other Tile Electronics (TRC, TCU, PCPU, RF power dividers & harness) are mounted on the back side of antenna. Typical conguration of a Tile and Panel is shown in Figure 28. The deck elements constitute conventional Pulse Doppler Radar, consisting
Fig. 29: Block diagram and photographs of deck elements

of two chains of receivers for simultaneous reception of V and H signals, one frequency generator, one Feeder SSPA and two sets of Data Acquisition and Compression units. The complete payload management is done by radar payload controller. Block diagram and photograph of deck elements are presented in Figure 29. Printed side of patch antenna side of the deployed antenna is shown in Figure 30. Figure 31 depicts RISAT SAR payload at different stages of testing and integration with spacecraft.

What is new in RISAT-1 SAR ?

RISAT-1, when conceived in way
Fig. 30: Patch antenna side of the deployed antenna

back 2002, was an ambitious project where many new technologies were required to be mastered. The access to these technologies was restricted because of international technology embargo, imposed on ISRO. The number of subsystems needed for this system was prohibitively large (precisely 1393 subsystems of them 312 are 8 bit computers). To qualify each of these subsystems to space grade, it required almost 3 weeks testing of each of these subsystems. ISROs experience with satellites never exceeded dealing with more than 100 subsystems. ISROs resources were inadequate to carry out fabrication and

Fig. 31: Clockwise from top left: RISAT SAR in testing; RISAT-1 in deployed condition backside and frontside; R ISAT-1 being lowered in thermovac chamber; RISAT-1 mounted a top PSLV

testing of such large number of elements. Indian industry was not equipped to handle

approach. A number of industries were needed to be hand-held, with training and expertise impartment to carry out mass productions based on nal blueprints which were designed and perfected in-house. The many rsts of RISAT-1 are: a. SAR Feature wise Only C-band SAR capable of giving 1 m resolution over 10 km x100 km spot. It is to be

5 - 10 km spot. It is the first SAR with Hybrid polarimetry. Linear polarimetric modes are available in some space-borne SAR systems, where data rate gets doubled and imaging is restricted below 35 incidence angle. This limitation is not imposed on Hybrid polarimetry and it is available seamlessly for all imaging modes. Further, this polarimetry mode can be self-calibrated using normal imaging data. It has a very low incidence angle modes operating at incidence angle as low as 12. It will usher in new application in soil moisture, glacier studies and better imaging in hilly regions. Apart from SAR, RISAT carries Indias first space-borne Active antenna array. The SAR payload, along with this antenna, is a complex array of electronics consisting of 1393 subsystems including 312 units of 8 bit computers. b. Payload Technology Completely indigenous Space qualified MMICs were fabricated in India Miniaturised TR modules Miniaturised EPCs with planar transformer technology where transformer windings are achieved by printing them on 16 layer PCB Programmable digital Chirp generators High speed digitization High data rate communication all the way upto 1.5 Gbits Down-linking data in two orthogonal circular polarizations with speed of 640 Mbits On-board recorder capability of 300 Gbits (240 Gbits at End of Life, EOL) Designed for truly Global operation

Industry Partnership
The highlight of RISAT-1 programme was development of hardware elements through industry partnership. Various new technology elements like different types of MMICs, miniaturized C-band TR-module and pulsed power supplies, dual polarized printed antenna, integration block and power distribution network, high speed digitizers, micro-controller and FPGA based central distributed control systems, etc. have been realized with the active participation and collaboration of public and 51

N N R M S

B U L L E T I N

mentioned that 1 meter SAR is presently possible with X-band only, that too over an area of

M A R C H

2 0 1 3

space quality fabrication, let alone handle such large numbers with zero defect production and testing

private sector industries like GAETEC, ASTRA, CENTUM, CMC, SHAHJANAND LASER etc. Indigenous MMIC fabrication line has been qualied at GAETEC foundry. Design, development and qualification of an indigenous Onboard Controller ASIC for Tile Electronics have also been accomplished in collaboration with private industries like CG COREL/AEROFLEX. The participating industries had to go through a learning curve with issues like space grade circuit fabrication, quality control, test facility development and testing methodology. Their contribution in mass production is
Fig. 32: Contributions of Indian industry in realisation of RISAT-1 SAR

illustrated in Figure 32. This programme has also added to industrial capacity building within the country. Thus, the challenge of microwave SAR payload realization has b e e n a d d re s s e d a n d w i t h i n d u s t r y participation in these activities, a new beginning has been made.

RISAT-1 Imaging Operations

Since 1st May, 2012, RISAT is being operated over India as well as over foreign countries. All the imaging modes except HRS have been operated successfully. Since
Fig. 33: Sample hybrid polarimetry images obtained by RISAT-1 in FRS-1 mode

October, 2012, calibrated images are made available operationally. Initial results indicate that system behavior is as per prediction and RISAT-1 system is expected to meet users expectations during its lifetime. A set of hybrid polarimetry images obtained in FRS-1 and MRS modes are presented in Figures 33 and 34. It is to be mentioned that, RISAT-1 is the rst SAR, capable of providing polarimetric images in scanSAR mode. D u a l _ F re q u e n c y S A R O n b o a rd Chandrayaan-2

Fig. 34: Sample hybrid polarimetry image obtained by RISAT-1 in MRS mode (Worlds rst polarimetry in scanSAR mode)

Indias second endeavour to moon,

conguration comprising Orbiter and Lander, to be launched in 2014. L & S-band Synthetic Aperture Radar (Dual-frequency SAR) is an important sensor in the instrument-suite planned for the orbiter. Dual-frequency (L & S-band) SAR has been envisaged to provide continuity to ongoing studies of S-band MiniSAR (onboard Chandrayaan-1) data and to
Fig. 35: Imaging geometry of Chandrayaan-2 SAR

The salient features of the L & S-band SAR are (Table 7): a. Congured for circular polarimetric mode and full polarimetric mode b. L-band SAR for greater depth of penetration in lunar regolith (which is about 5-10 m, depending on FeO+TiO2 content of lunar regolith) c. Wide range of resolution options (2 m 75 m slant range) d. Wide range of incidence angle coverage (9.5 - 35). (Figure 35) e. L-band and S-band SAR operation in stand-alone or simultaneous mode of operation f. Miniaturized hardware to meet 14.3 kg mass constraint The basic conguration of Chandrayaan-2 SAR payload is shown in Figure 36a. DVM development of the complete SAR system is already completed and is shown in Figure 36b. FM development of the payload is targeted to be over by Dec-2013. Table 7: Specications of Chandrayan-2 SAR Frequency 1.25GHz (L-band), 2.5GHz (S-band) Circular and Full-Polarimetry 75MHz to 2MHz selectable 10 km 2 m 75 m selectable Antenna Microstrip patch antenna of 1.35 m x 1.1 m of gain 22 dBi (L-band) and 25 dBi(S-band) 45W (L-band), 40W (S-band) 24% (max) 160 Mbps (max, combined H&V channels)

Polarization Bandwidth Swath Resolution

Peak power Duty Cycle Data Rate

Raw Bus Power 100W (max)

Sigma-26 dB (L-band-Circular) Naught @2m -22 dB (S-band-Circular) resolution L-Band SAR Satellite Presently plans are afoot to build an L-band SAR as successor of RISAT-1. Presently the payload is in drawing board. This SAR will have scanSAR modes matching with RISAT-1. It will have strip-map mode of resolution 5 m and swath of 60 m. The interesting addition will be electronically steered spotlight 53

N N R M S

cater to further advanced scientic studies.

B U L L E T I N

M A R C H

2 0 1 3

Chandrayaan-2, is planned as a two-module

mode providing 2.5 m resolution over 40 km x 40 km spot. This SAR will also provide moving target detection through Along Track Interferometry (ATI).Compared to RISAT-1, it will have additional linear polarimetry mode apart from Hybrid polarimetry. Basic imaging modes are shown in Figure
Fig. 36: (a) Overall L/S-band SAR system conguration (b) DVM of L/S-band SAR system

37 and the performance specications are presented in Table-8. Like RISAT-1, the antenna will be based on active array technology to provide electronic steering in elevation and limited electronic steering in azimuth to support spotlight mode. The antenna size will be 10 m x 3 m, weighing 600 kg. and the satellite is being planned for launch by PSLV. L-band SAR pay load will also carry a demonstration model of real time processor, for the rst time, for processing the SAR raw data on the y. The mechanical conguration of the proposed L-band SAR system is shown in Figure 38.

Fig. 37: Imaging Conguration of L-band SAR

Development of Building Blocks for Microwave Remote Sensors

Signicant development of range of microwave remote sensors in ISRO, was possible because of gradual acquisition of key
Fig. 38: Mechanical conguration of L-band SAR with satellite structure

technologies in antennas, RF receivers, power

Table 8: Salient Specications of L-band SAR 1 2 3 4 5 6 7 8 Frequency Polarization Swath Range Coverage Incidence Angles Spatial Resolution Minimum Sigma Naught Radiometric Resolution L-BAND (1.25 GHz) Single, Dual, Circular (Hybrid) Polarization and Full-polarimetry 50 km To 200 km Between Off-nadir Distance Of 150 km and 500 km ~15deg To ~45deg 5 m, 10 m & 25 m -20/-23 dB (multilook) Or Better 2-3 dB

Whether passive or active sensors, very low noise receivers are must component of microwave sensors. Presently expertise exists in ISRO for development of receiver all the way from UHF to G band (183 GHz) (Figure 39a). This expertise development called

and high precision fabrication facilities. Off late, emphasis has been given to acquire expertise
Fig. 39: Imaging geometry of Chandrayaan-2 SAR

in Monolithic Microwave Integrated Circuit (MMIC) technology (Figure 39b). Presently design capability exists in designing MMICs upto V band. Microwave Remote Sensors call for large number of complex RF receivers. This capability is not only important from the point of view of miniaturizing the receivers but also important from strategic aspects. For active sensors, Solid State Power

Fig. 39: (a) Range of receiver expertise acquired under MRS activity (b) Range of MMIC expertise acquired under MRS activity

Amplifier (SSPA) technology is gradually replacing costly Travelling Wave Tube Amplier (TWTA) technology. Consequently a strategic initiative has been taken to master SSPA designs, from a few Watts to a few hundred Watts. Also graduation has been done from low efciency GaAs technology to high efciency GaN technology. Similarly SSPA circuit topology has migrated from classical class AB to very sophisticated class F. Presently SSPA

Fig. 39: (c) Range of SSPA expertise acquired under MRS activity (d) Range of EPC expertise acquired under MRS activity

design and development capability exists from L-Ku band and efforts are on to migrate to

higher bands like Ka band and beyond (Figure 39c). EPCs are key to building successful systems with optimum performance and high longevity. The array of expertise in EPC technology, under MRS activity, ranges from low power-high voltage to low voltage-high power to pulsed operation. Gradually our EPC designers migrated from Printed Circuit Boards (PCBs) to Hybrid Microwave Circuit (HMCs), LDOs to POLs, coiled transformers to planar transformers (Figure 39d). The key drivers are miniaturization and seamless integration within targeted Radio Frequency (RF) or digital packages to reduce Electro Magnetic Interference (EMI) / Electro Magnetic Compatability (EMC) effects. 55

N N R M S

B U L L E T I N

for large investment in different design, test

M A R C H

2 0 1 3

amplifiers, EPCs and digital subsystems.

As far as digital sub-systems are concerned, discrete devices are passive. A strong expertise have been developed in designing programmable and recongurable digital systems using high end fault tolerant Field Programmable Gate Array (FPGAs). Already migration from FPGAs to Application Specic Integrated Circuit (ASICs) has happened and present effort is there for designing and manufacturing a large range of ASICs as key to miniaturization of technology. Similarly already we have expertise for handling data digitization and modulated signal generation up-to 700 MHz and efforts are on to stretch the limit beyond 1 GHz within a short future (Figure 39e). All the efforts have now resulted in much miniaturized and low power digital subsystems.

Conclusion
Last two decades have been invested in understanding the nuances of Microwave
Fig. 39: (e) Improvement in the performance of digital subsystems under MRS activity

Remote Sensing technology, both in terms of system development and algorithm. It is

demonstrated in successful journey from SAMIR to RISAT. However, there are three formidable challenges: (i) Ability to master a diverse array of technologies which function with precision. The range of frequencies encompassed today ranges from 1 GHz to 183 GHz. Complete spectrum of devices change drastically over these bands. Each of the bands provide different challenges in terms of design, simulation, fabrication and testing requirements. Consequently wide investments are to be made in the all the above elds. Coupled with it, they pose complexities in system designs to varying degrees. As an example TSU, where lters are to be achieved are to the order of 10 MHz at 50-60 GHz bandwidth. The estimate of emission is to be made to the accuracy of 0.5 K. This calls for high degree of machining of RF fabrication requirements. The testing system has to be evolved as conventional test systems will not meet these requirements. Take another example of RISAT consisting of close to 1400 sub systems, which includes 312 computers. The complexity is equivalent to fabricating 5 INSAT satellites. The sheer number of subsystems forced us to look towards harnessing Indian industries on partnership basis. But Indian industry scenario is not geared to high quality fabrication required for space systems. The strategy included (a) convincing industries that they can do it (b) they had to be hand-held during their initiation with tough fabrication methodology and processes (c) resorting to partnership with industry in place of treating them as any other vendor. (ii) A great emphasis is to be provided to algorithm development for testing or processing or simulation. In fact all the designers need to be trained in simulation and algorithm development. In DMSAR, a completely new SAR processing

the sheer requirement of measuring active antenna pattern at different stages of integration of RISAT-1, led to the development of a time domain near eld antenna pattern measurement facility in the clean room itself. The need for calibrating the active antenna led to development of a single scan calibration methodology. Fear of accidental damage of RISAT-1 payload by external AC supply, led to the development of a concept of a payload measuring itself. Before introduction of Hybrid polarimetry, as a new concept in RISAT-1, tremendous simulation activity had to be undertaken to explore whether there is any lacunae in the concept. (iii) The microwave signatures do not conform remotely to human experience with vision. Interpretation of signatures to physical reality needs considerable understanding of the physics of interaction of microwave signals to the physical objects. A great deal of analytical and experimental studies are required to nally bring the fruits of microwave remote sensing technology to the last mile post : the users.

Acknowledgement
Development of Microwave Remote Sensing technology, in all its gamut, would not have been possible without active contributions and participations from a large number of engineers and scientists who worked as a team. Author would like to take this opportunity, to acknowledge and salute the individual team members who made this journey successful.

57

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

algorithm had to be developed for severe motion errors and lack of stabilization system. Similarly,

AGRICULtURAL APPLICAtIONS: EvOLUtION DURING LASt 25 YEARS

Parihar JS and Manjunath KR Space Applications Centre, (ISRO), Ahmedabad 380 115, India Email: jsparihar@sac.isro.gov.in

Introduction
India has a geographical area of 329 mha with a wide diversity of its natural resources spread over 15 agro-climatic zones. Out of which, the agricultural area is 142 mha with 37 per cent of the area under irrigation and about 215 crops are grown in India. The diverse crop growing conditions coupled with uncertainties of climate situation demands crop information need on temporal and spatial scale. Though India receives 420 mha m of rain and snowfall every year, still is prone to oods and droughts due to uneven distribution. Apart from this, it is prone to several other natural disasters. In view of these, the country needs accurate and timely information on the agricultural resources at regular intervals. Realizing the signicance of orbital remote sensing, country initiated, evolved and developed its own space programme for earth observations. The agricultural remote sensing during the last twenty five years has reached high level of maturity in various domains. Consistent efforts have been made on exploring the Remote Sensing (RS) applications, technique development, user interaction and institutionalisation. The following sections explain the developments taken place during the last three decades with special reference to agricultural remote sensing in India.

Crop Production Forecasting

Remote sensing (RS) technology has potential in estimation of crop area and forecast productivity at district and regional level due to its multispectral, large area and repetitive coverage. Exploring the use of remote sensing for agricultural application in India started with the use of multi band and Colour Infra Red (CIR) aerial photographs as early as 1974-75 (Sahai et al., 1977). Further knowledge on crop signature was gathered through scientically designed eld experiments using multi band radiometer under the Indian Remote sensing Satellite-Utilisation Programme (IRS-UP). Since then a large number of experiments have been carried out for developing techniques for extracting agriculture-related information from airborne and space borne data. Many of these studies have led to the operationalisation of the methodology and conduct national-level projects. Crop production forecasting comprises identication of crops, acreage estimation and forecasting their yield. Crop identication and discrimination is based upon the fact that each crop has a unique reectance pattern in different parts of the electromagnetic spectrum which is termed as spectral signature. The general spectral response of a crop canopy in the visible and NIR region is characterised by absorption in the 0.35 to 0.5 um and 0.6 to 0.7 m regions (due to chlorophyll pigments), high reectance in the green region (around 0.54 m), a steep increase in the reection in the 0.7 to 0.74 m and high reectance in 0.74 to 1.3 m region due to internal cellular structure

The varying response of the crops stems from the fact that various factors such as type of crop, stage of the crop, canopy architecture, per cent ground cover, differences in cultural practices, crop stress conditions, background soil/water etc., contribute to the composite response. Each crop has its own architecture, growing period, etc. thus enabling discrimination through remote sensing data. If there are two crops with similar spectral signatures on a given date (confusion crops), multidate data are used to discriminate them. Vigour of the crop is manifested in the absorption in near infrared to red radiance is a good indicator of the vigour of the crop. Some of these properties are utilised in crop identication, crop condition assessment and yield forecasting. Launch of the Indian Remote Sensing Satellites (IRS-1A & 1B) carrying Linear Imaging Selfscanning Sensors (LISS I & II) in 1988 and 1991 provided a much-required impetus to agricultural applications. Further boost was provided by the state-of-the-art satellite IRS-1C in December 1995 and IRS-1D in 1997. These satellites carried onboard a unique combination of three sensors viz., (i) Wide Field Sensor (WiFS) with 188m spatial, two spectral bands red and near infrared, 810 km swath and a repetivity of 5 days, (ii) Linear Imaging Self scanning Sensor (LISS-III) with 23.5 m spatial resolution in the green red and near infrared region, and 70.5 m in the middle infrared region, and 140 km swath, and (iii) Panchromatic (PAN) camera with 5.8 m spatial resolution, 70 km swath and stereo capability. IRS-P3 MOS provided data in many narrow spectral bands, although at coarse spatial resolution facilitating the study of crop stress and crop senescence. The launch of RISAT has lled another dimension to the agricultural remote sensing as all weather capability of data is now reality from Indian satellite. Use of multi band and Colour Infra Red (CIR) aerial photographs was explored since inception of remote sensing applications programme in the country. Projects on I) Crop production forecasting ii) Crop stress detection, and iii) Crop yield modeling were taken up in 1983. Early success in these, led to launching of Large Area Crop Acreage (LACA) estimation experiment in 1986. Regional level wheat acreage estimation using Landsat MSS data was demonstrated (Sahai and Dadhwal, 1990, Navalgund et al., 1991). Results of LACA were appreciated by the Ministry of Agriculture in the country and at its behest Crop Acreage and Production Estimation (CAPE) project was launched in 1988 (SAC 1990). This project was sponsored by the Ministry of Agriculture, was executed jointly by Department of Space, State Remote Sensing Centres, State Departments of Agriculture and Agricultural Universities. Pre-harvest production estimates carried out are of two types one: small scale (district-level, single forecast) and two: regional scale (National level, multiple forecasts). Pre-harvesting production forecast of wheat, rice, cotton, rapeseed/mustard, groundnut and winter sorghum is made under the project. Remote sensing data of moderate spatial resolution (Landsat MSS & TM, IRS LISS I, LISS II & LISS III) acquired at optimum bio-window was used for crop area estimation. The data was analysed applying Maximum Likelihood Supervised classication technique, where limited eld information called Ground truth is used to generate the training signature. In case of complete enumeration data for analysis was selected by overlaying the boundary mask of the area over the remote sensing images. When it was extended to large area stratied sampling technique was developed where area was rst divided with a grid representing sampling frame size (5 X 5 km) and then data belonging to selected sample segments is extracted (20%) are analysed. The samples are randomly drawn the red and reectance in the near infrared spectral regions. It has been observed that the ratio of

59

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

of the leaves. The absorption at 1.45, 1.95 and 2.6 m spectral bands is due to leaf water content.

proportionate to size of each stratum. Area estimate is made from the proportion of crop present in the sample. The CAPE project was further followed by CAPE-II, second phase. A semi-automated procedure called CAPEWORKS was developed for this purpose. The scope of the CAPE project progressively increased since 1988 to include more crops as well as multiple
Fig. 1: Information need and sources in frequent crop monitoring

forecasts as the crop season progresses. Since the space technology has advanced and variety of sensors of different spatial, spectral and temporal resolutions are available and there is a continual need for crop information throughout the growing period (Figure 1), frequent monitoring is feasible at various scales. Realizing the importance of multiple source information like weather, econometric and eld survey towards a robust approach for multiple forecasts of a number of crops, a new concept Forecasting Agricultural output using Space, Agro- meteorology and Land based observations (FASAL), was formulated by Parihar and Dadhwal (2002) (Figure 2). Implementation of FASAL was initiated in 2007-08 and SAC was entrusted with the responsibility of implementing Space Technology based production forecast of crops and upgradation of the procedure with new data availability. Thus, the work components encompass two aspects: operational forecast of production and developmental aspects. During the 11 th Five Year Plan (2007-8 to 2011-12) procedures for multiple forecasts of 6 crops viz: jute, kharif rice, winter potato, rapeseed/mustard, wheat and

Fig. 1: Concept of FASAL

Fig. 3: Time line of activities leading to the formation of MNCFC

rabi rice was developed and regular estimates were made available. Realising the need to

integrate this advanced technique in the routine crop statistics gathering, Department of Agriculture and Cooperation (DAC), Ministry of Agriculture, Govt. of India initiated steps to set up a centre for this purpose. Accordingly, the centre

Delhi on April 23rd, 2012. Key features of FASAL are (a) Institutionalising the operational use of RS data for diverse applications in agriculture, (b) Developing a system for multiple in-season crop assessment and forecasting in near real-time, and (c) Integrated use of tools such as GIS, large databases, modelling and networking. The concept of FASAL thus strengthens the current capabilities of early season crop estimation from econometric and weather-based techniques by adding the tools of remote sensing in a major way. A timeline of activities leading to the formation of MNCFC is shown in Figure 3. Currently MNCFC covers National forecast of kharif Rice: 13 states (90%area), rabi rice:4 states (95% of area), wheat: 7 states (95% of area), jute: 3 states (99% of area), mustard: 5 states (84% of area) and winter potato: 4 states (88% of area). Customized software called FASALSOFT was developed at Space Applications Centre, ISRO, Ahmedabad to cater to these applications.

National-level wheat acreage estimation using multi-date WiFS data from IRS-1C was initiated in 1995-96 season. Multiple in-season wheat forecasts using multi-date IRS WiFS data and weekly weather variables at meteorological subdivisions in major wheat producing states of India are being made at national level. Seven major wheat-producing states, Uttar Pradesh, Punjab, Madhya Pradesh, Haryana, Rajasthan, Gujarat and Bihar form the study area. These states account for more than 90% of the wheat production. The sample segments are classified using in-season ground truth and a hierarchical (decision rule based) classier. The state level acreage estimates are then statistically aggregated
Fig. 4a: The rainfed (left region) and irrigated (lower right) wheat in Madhya Pradesh

to arrive at national level wheat acreage estimates. The methodology for state level yield forecasting is multiple regression models based on temperature, using a correlation weighted regression approach. The National level acreage and yield estimates are then combined to provide National Production Forecast (Oza, et al., 2002). The wheat crop under rainfed and irrigated regions of MP and inter-annual changes between 2011 and 2012, showing

Fig. 4b: Inter-annual changes- Increase in wheat crop (Kota, Rajasthan

variation of wheat crop is shown in Figures 4a & b, respectively.

FASAL: National Rice Production Forecasting

Investigations using space borne SAR data started with limited use of data from JERS, SIR C sensors. However, the possibility of examining space borne radar data for large area agricultural application was realised with the successful launch of ERS-1 Synthetic Aperture Radar (SAR) (SAC, 1997). Due to the problem of cloudy weather during rainy season, kharif rice crop production 61

N N R M S

FASAL: National Wheat Production Forecast

B U L L E T I N

M A R C H

2 0 1 3

Mahalanobis National Crop Forecast Centre (MNCFC) has been set up at PUSA, Rajendranagr, New

estimates in the major rice growing states are being generated using multi-date Synthetic Aperture Radar data (Figure 5a, Patel et al., 2004). Rice growing environment or management practices ensures that there is standing water beneath the canopy at least for a short duration during crop season. This
Fig. 5: (a) Three date color composite of ScanSAR data ( R:G:B: Date1:Date2: Date 3) over the study district showing distinct signature of rice (b,c,d,e,f) in different growth stage and other land cover classes (a: water, h: forest and i: urban). (b) Temporal pattern of rice crop

information is used to characterize the rice crop on a temporal domain (Figure 5b). The rate of change and direction of change of SAR response aids in building decision rule for classication of rice pixels. A stratied random sampling approach for each state is adopted for acreage estimation with a sample size of 5 x 5 km. A fteen per cent sampling fraction and in-season ground truth information of the selected sample segments

Fig. 5c: Multidate SAR data showing decrease in kharif rice crop (Begusarai, Samastipur, Bihar) in 2011 in comparison to 2012

are used. The segments are classied using a decision rule classier followed by statistical aggregation of state level acreages to national

level rice acreage estimate. The statistical relationship between yield and rainfall during the cropping season is used for yield forecasts. The district level models are combined with acreage and production forecasts for the country is made. This is being regularly carried out since 1998. The inter-annual changes are shown in Figure 5c.

Assessment of Crop Damage and Crop Progress

Damage to crop due to moisture stress is a common occurrence in rain fed rice growing region. The characteristic backscatter prole of rice using temporal SAR data is useful in characterising the crop condition as normal and sub-normal. Flood is a common phenomenon in many rice-growing regions, particularly in monsoon season. Temporal SAR data is found not only to map ood affected rice elds, but also to model duration and degree of submergence. Complete submergence of rice at any given period of growth lower the backscatter. The degree of submergence was modelled with reference to crop height and its deviation from the reference normal growth prole. The model thus can detect the completely submerged elds as well as partially submerged elds. Crop planting depends on many factors. There is a latitudinal variation in the rice crop calendar. This gives rise to a wide period in crop planting with in any country/region. Spatial crop calendars can be of great advantage to study such processes. Specic advantage of SAR sensitivity to target properties were used to categorise the rice area based on the sowing period. It is well established that sowing dates have a signicant effect on crop biomass and yield. Temporal SAR data is used to categorise the elds as normal, late and very
Fig. 6: Rice spectral prole of different regions in India derived using multidate optical data

late sown which is additional component of

the reasons for delay. Efforts have been made for deriving the regional variation of rice growth prole using optical data (Figure 6; Manjunath et al., 2006)

Crop Yield Models

Crop yield assessment is complex growth processes. Every crop genotype has certain yield potential, which can be achieved (to an approximation) in experimental eld with optimal conditions. However, in the real world, the crop yield is conditioned by various abiotic (soil, weather and cultivation
Fig. 7: Relationship between rice biomass and SAR backscatter

practices, like date of sowing, irrigation and fertilizer) and biotic stresses (disease and

pest). While the variability of the weather explain most of the annual variability over a short period of time, the cultivation practices and new varieties explain most of the variability over a period of 10 to 20 years. All these factors are interdependent and make the yield assessment a more complex task. Hence, one way of forecasting the yield is accounting the variability in the above parameters and dening their relationship with the nal crop yield. Satellite based remote sensing provides a suitable alternative for crop condition and yield assessment/ forecasting, as it gives a timely, accurate, and synoptic estimation of various crop parameters. The time series based trend and arima models were developed in the beginning of CAPE project based on district-wise yield data of DES, which were used to compute production by multiplying with the RS derived area estimates (Agarwal et al., 1983). Further agrometeolorlogical model, spectral models and combination of these models were tested and used for a variety of conditions, crops and regions. The rice biomass has high correlation with SAR backscatter (Figure 7) and therefore the yield. Hence, SAR backscatter estimation using radiative transfer and estimation of rice biomass and yield using its inverse is now standardized. The vigour of crop, an indication of yield is being assessed using vegetation indices derived from different parts of the spectrum. The normalized vegetation index is one such index which represents the green biomass of the plant and was directly correlated to the yield of the crop and this relation was used for estimating yield ( Manjunath et al., 2002). A variety of models involving combination of factors of weather and spectral parameters have been developed and used in conjugation with remote sensing derived acreage for providing production estimates. Plant growth simulation models have been used for monitoring crop growth and predicting yield. However, their use in large areas has been limited because most plant growth models were developed at the eld scales and the performance of the models is not so satisfactory when they are extended from eld to regional scales due to the unavailability of crop information at regional scale. The main advantage of remotely sensed information is that it provides a quantication of actual state of the plant on a regional scale and shows the spatial variability. While models provide a continuous description of plant growth over the period of interest, remotely sensed observations are discrete time events. Hence integrating the inputs from remote sensing to the crop models is useful technique to predict crop yield at large scale. Leaf Area Index (LAI) is most important in explaining the ability of the crop to intercept solar energy and in understanding the 63

N N R M S

B U L L E T I N

because of the high variability involved in the

M A R C H

2 0 1 3

crop assessment need that enables identifying

impact of crop management practices. Hence spatial distribution of LAI was derived using NDVI data and wheat yield forecasting at spatial scale was tried through forcing the LAI in a crop simulation model WOFOST. As the model is point based, its spatial application required spatial database generation (both agro-meteorological and biophysical) and linking of the spatial database to the WOFOST application program. The broad procedure is shown in Figure 8. Using this procedure in-season wheat yield forecasting at 5 km 5 km has been demonstrated (Tripathy et al., 2013).
Fig. 8: The crop simulation modeling procedure using RS data

With the availability of inputs like insolation from Kalpana satellite, the renement of this approach would now be feasible (Figure 9). A lot of experience has been gained for yield model development using space-borne RS data directly (spectral yield model) and their use in pre-harvest yield forecasting for wheat, rice, cotton, mustard and groundnut. The approach followed has been direct regression of spectral indices (NIR/R of NDVI) to district level yields. The crop yield data used is in two format (i)
Fig. 9: The yield variation pattern obtained using interpolated insolation and Kalpana derived insolation

site specic yields where the Crop Cutting Experiments (CCE) are specically carried out (ii) those collected by State Department of Agriculture as a part of yield estimation programme. These are used to derive district average yield, which are then related to RS based indices.

Kharif Progress Assessment

The technique of early assessment of crop prospects based on soil moisture modeling using Satellite-based daily spatial rainfall has been developed during last ve years. Early assessment of All-India kharif rice crop acreage (progress of sowing) for the year 2012 was carried out at 7-day interval
Fig. 10a: Kharif rice area sown based on Available Soil Moisture (ASM)

starting from July 15, with nal assessment on

maps showing various parameters like available soil moisture, average irrigation requirement, aridity index for the current year, last year (2011) and last 5-year average (2007-2011), aridity anomaly and current year soil-moisture rank were generated. The colour-coded maps show, at a glance, agricultural prospects at 7-day interval as the kharif season progresses.

Rabi Progress Assessment using INSAT 3A CCD

Vegetation mapping using satellite data has became a regular practice in recent year. The advancement in satellite technology and improvement in the resolutions of data for monitoring the vegetation growth is carried out all over the world using high resolution satellite data. This attempt to monitor the progress of rabi crop using the Indian Geostationary Satellite is rst of its
Fig. 10b: Rabi crop progression derived using INSAT CCD

kind. The rabi season NDVI proles were extracted and decision rule were made by

applying proper NDVI threshold. For each decade, decision tree has been updated according to NDVI value. Following this methodology, rabi progress has been assessed at each decade during the year 2011-12 and 2012-13 (Figure 10b, Vyas et al., 2011)

Remote Sensing Applications in Horticulture

India is bestowed with varied agro-climate which is highly favorable for growing a large number of horticultural crops such as fruits, vegetables, root tuber, ornamental, aromatic plants, medicinal, spices and plantation crops like coconut, areca nut, cashew and cocoa. India is the largest producer of fruits (49.36 MT) and second largest producer of vegetables (93 MT) in the world. Horticulture occupies about 12 per cent of the total cultivated area in the country, and contributes about 25 per cent of the total agricultural export. Remote sensing technology helps in generation of crop Inventory of major horticultural crops, site suitability analysis for expansion/introduction, infra structure planning for post harvest requirement, disease detection and precision planning for horticulture. The general approach involves the use of high resolution/high temporal data (LISS-III) for identifying the crop of interest and relevant collateral information (i.e soil, water, climatic, infrastructure etc) and processing for logical clustering for decision-making. Feasibility studies demonstrating the remote sensing technology in horticultural sector have been carried out. Inventory of orchards like apple (Figure 11), grape, mango, coconut, banana and vegetables like potato, onion has been carried out in different agroclimatic regions of the country (Panigrahy and Manjunath, 2009). The post-harvest infrastructure planning and optimization of cold store facility for post-harvest management of
Fig. 11: Apple orchards as seen in LISS-III FCC and classied image

potato has been demonstrated (Panigrahy

65

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

September 30, 2012 (Figure 10a). Along-with statistics of rice acreage, various all-India colour-coded

and Manjunath, 2009). Apart from these early trends, in national winter potato production from the country is regularly brought out to infer about National/state production prospects and identifying areas with signicant change. As a part of Technology Mission for Integrated Mission of Horticulture in North Eastern states including Sikkim, remote sensing based studies in the following states has been carried out (Arunachal Pradesh: Apple; Meghalaya: Cashewnut; Tripura: Pineapple; Mizoram: Passion fruit cultivation
Fig. 12: The LISS-IV FCC showing onion crop elds

in zoom area; Manipur: Pineapple; Sikkim: Mandarin; Himachal Pradesh: Apple). The latest addition is onion crop estimation, which uses a combination of LISS-III and LISS-IV (Figure 12) and methodology was transferred to NHRDF, Nashik which makes periodical estimation of late kharif and rabi onion with technical support from SAC.

Agro-Ecosystems Assessment and Modelling

During the last two decades techniques of deriving and assessing the agroecosystems components were developed which include cropping systems analysis, methane emission inventory, parameter retrieval and process modelling, energy and mass exchange in vegetative systems etc. Some of them are described below:

Cropping Systems Analysis

A cropping system is dened as the cropping pattern and its management to derive benets from a given resource base under a specic environmental condition. This requires identification of crops and areas where changes in cropping patterns are desirable. This calls for an initial step of creating an updated database of the present cropping systems of the country and simulate the long-term effects, taking into consideration the resource base and agroclimatic condition. Satellite remote
Fig. 13: The cropping pattern and crop rotation maps derived using RS data

sensing (RS) and Geographical Information

seasonal cropping pattern, sowing pattern, crop rotation, efciency indicators and other related parameters. The pilot studies on cropping systems was initiated by SAC in Punjab and West Bengal states, which represent high potential and high productivity and high potential and low productivity conditions, respectively. The multidate SAR data was used deriving the kharif season cropping pattern and multidate WiFS/AWiFS data was used for deriving rabi and summer season pattern. These were logically combined to derive the crop rotation maps (Panigrahy et al., 2010). Encouraged by the ndings, the study was extended to the entire Indo-gangetic plains (Figure 13). The outputs generated in the study were seasonal cropping pattern, crop rotation, kharif fallow area maps as well as crop intensication and diversity maps. The cropping systems performance indicators were also generated. The Multiple Cropping Index (MCI), Diversity Index (DI), and Cultivated Land Utilisation Index (CLUI) have been deduced using remote sensing data (Manjunath et al., 2011). An illustration of West Bengal is shown in Figure 14. The crop diversification aids in efficient management of resources, maintaining the soil fertility and reduced pests/ disease incidence. As part of pilot study the spatial map of crop diversication of Punjab state was prepared (Panigrahy et al., 2010a). Similarly the techniques of crop intensication based on cropping pattern, rotation, fallow area/duration and invariant resources was developed in the above study.

Environmental Impact Assessment of Agricultural System

The agriculture has transformed from simple sustenance objective to intensiveFig. 14: Cropping system performance indicators

commercial form thus depleting and degrading the environmental resources. Pressure on high

production has led to intensication of agriculture. Intensive Agriculture, long term sustainability and quality of natural resources, thus is matter of compromise and concern. Agricultural is a major reservoir and transformer in global cycles of carbon, nitrogen and water. Intensive agriculture leads to erosion of soil resources, loss of biodiversity, alienation of ecological niches, temporary imbalance in soil microbial functioning, associated long-term effects on microbial processes and changes in biogeochemical cycles. Agricultural activities contribute about 70% of all anthropogenic N2O emissions and about 65% of all anthropogenic CH4 emissions. Nutrient leakage from agriculture is a prime cause of degradation of groundwater, surface waters and estuarine and coastal marine systems, and via the atmosphere affects 67

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

System (GIS) have a crucial role to play in this direction. The multidate satellite data is helpful in deriving

other terrestrial systems. Nitrate contamination of groundwater is common in agricultural areas around the world. Some of the specic components include fertilizer and pesticide residual toxicity, plant/soil metabolic exudates such as methane/ nitrous oxide in the immediate micro/macro environment. Methane and nitrous oxide form important components of such an interface. The issue related to methane emission from rice elds and livestock from India is addressed in detail in the project entitled Environmental Impact Assessment in Agriculture System. The GHG pattern which is also available using sensors on board satellites is being studied in detail (Panigrahy et al., 2010). Apart from this attempt has been made to grossly quantify the nitrous oxide emissions from live stock. The environmental impacts need spatio-temporal assessment with best possible scientic tools and techniques. Use of Remote sensing and GIS along with eld and collateral data helps in deriving and assessing the important parameters required for environmental impact assessment. According to IPCC guidelines, the rice ecosystem need to be categorized into four strata for methane emission study. Ancillary data on rainfall, elevation, soil, command area/ irrigation statistics were used with remote sensing derived rice maps in GIS to categorize the rice lands into four strata as: irrigated, rain fed ood prone and rain fed drought prone and others (Manjunath et al., 2006). In-situ weekly measurements of methane emission from the representative ecosystems was collected and analysed using gas chromatography following the IPCC standards for three consecutive years; 2003, 2004 and 2005. The methodology was developed to generate total annual methane emission map from the rice areas of India and its temporal pattern taking into consideration the diverse conditions under which the rice is grown (Manjunath et al., 2011). The methodology was developed for the variety components on use of RS data for stratication, spatial and temporal sampling strategy, development of indigenous method of sampling and upscaling of methane. Livestock constitutes an integral component of Indian agriculture, which is another major source of methane emission mostly from enteric fermentation by ruminants. A detailed state/ district-level methane emission inventory for different livestock categories was made using the country-specic and Indian feed standard based methane emission coefcients, which are based on IPCC guidelines, and the latest available livestock census. The total methane emission including enteric fermentation and manure management has been estimated as 11.75 Tg for the year 2003 (Chhabra et al., 2008, 2009).

Non-point Source Pollution

Agriculture has been identied as the largest contributor of non-point source
Fig. 15: An illustration of non-point source pollution studies using remote sensing and eld data.

(NPS) pollution of surface and ground water systems globally. The non-point

was also studied using remote sensing derived inputs in GIS environment (Chhabra et al., 2010). Fertilizers, which are used as important inputs in agriculture to supply essential nutrients like nitrogen (N), phosphorus (P), and potassium (K) also, serve as a major non-point source pollutant. An integrated methodology was developed for quantification of different forms of nitrogen losses from rice crop using remote sensing derived inputs, eld data of fertilizer application, collateral data of soil and rainfall and nitrogen loss coefcients derived from published nitrogen dynamics (Figure 15) of kharif and rabi seasons. The nitrogen losses through leaching in form of The study results indicate that nitrogen loss through leaching in kharif and rabi rice is of the order of 34.9% and 39.8% of the applied nitrogenous fertilizer in the Indo-Gangetic plain region. This study provides a signicant insight to the role of nitrogenous fertilizer as a major non-point source pollutant from agriculture. urea-N, ammonium-N (NH4-N) and nitrate-N (NO3-N) are dominant over ammonia volatilization loss.

Impact of Climate Change on Agricultural System

Climate change is one of the most discussed topics of the last two decades. Its impact on major agricultural systems of India with inputs from multiple sources was studied using simulation model CropSyst (Tripathy et al., 2011). The climate forecasts from different GCMs, RCM and statistical downscaled model (under different scenario) were used for understanding the impact of climate changes on agricultural systems. In addition to the impact of mean seasonal climate change, index based analysis was carried out for the past weather extremes. Trends of extreme weather parameters under future climate change scenarios were derived and their impact on crop productivity was studied. The study also identied optimum planting date as one of the adaptation strategies for the agriculture sector to cope with the projected impacts in the study regions and analyzed the uncertainty due to both climate model and impact model.

Energy and Mass Exchange in Agro-ecosystems

An internalized calibrated model and evaporative fraction-based model were used to estimate large-area sensible and latent heat uxes over semi-arid climate of Gujarat. The validation with collocated measurements over rice-wheat system from INSAT-linked Agro-Met Station (AMS) and Large Aperture Scintillometer (LAS) showed improvement in validation accuracy when area-averaged measurements from LAS were used as compared to AMS measurements monthly sensible heat uxes were found to have generalized, inverse and strong power t relation with rice and wheat grain yield. Regional climatology of evapo-transpiration was generated using 30 years (1981 - 2010) satellite based long term optical and thermal band observations, upscaling functions for energy balance components from measurements of INSAT-linked AMS network. Mann-Kendal test statistics showed Evapo Transpiration (ET) change hotspots over India. Dekadal ET-change during 1991-2000 and 20012010 showed opposite change (+ve to ve or vice versa) behaviour. However, majority part of India including central, north-west, part of eastern and northern India showed a declining ET trend during 2001-2010 which resemble with trend in rainfall and surface soil moisture in majority patches. The climatology from 30 years showed a range of annual evapo-transpiration 100 to 1300 mm. Surface insolation product using Indian geostationary satellite (Kalpana-1) was made operational. Relative insolation was used to determine monthly and annual frequency of clear days and assured solar energy exposure (Figure 16) over India on monthly and 69

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

source pollution of nitrogenous fertilizers in the Indo-gangetic plains vis--vis fertilizer consumption

annual scale using three years of datasets (2009 to 2011). It showed well differentiated zones of low (500 MJm-2) to high (4000 MJm-2) solar energy zones.

Biophysical Parameter Retrieval and Process Modeling

The satellite derived biophysical products is one of the key developments taken place during last two decades. The investigations on deriving these products were carried out in India using Indian and other sensors. The NDVI, FAPAR, insolation,
Fig. 16: Assured solar energy potential over India from Kalpana-1 insolation product

LAI, LST etc are some of them on which R & D were carried out. The Leaf Area Index (LAI) is a key biophysical variable used by plant physiologists and modelers for estimating foliage cover and plant growth and biomass. The regional modeling of growth processes such as evapotranspiration (ET) and net CO2 assimilation require retrieval of some core variables such as land surface temperature (LST), leaf area index (LAI), albedo and soil moisture etc.. Field-scale (local) and regionalscale (agro-climatic zone) non-linear empirical models are developed for wheat leaf area index (LAI) based on normalized difference vegetation index (NDVI) using IRS 1D LISS-III

Fig. 17: Spatial distribution of maximum Leaf Area Index for wheat crop from daily INSAT 3A CCD data for year 2011-12

(Chourasia et al., 2006) and IRS-P6 AWiFS (Chourasia et al., 2011) data. The modeled wheat LAI were validated with independent ground measurements through systematic campaigns and the range of errors was found to be 0.7-1.2 with higher errors at LAI more than 3.0 due to saturation of NDVI. The atmosphere-corrected reectance of AWiFS was generated using SMAC (Simple model for Atmospheric Correction) using real time MODIS atmospheric products of aerosol, ozone and water vapour. These atmospheric corrected AWiFS reectance were used to invert the ProSail model through look up table to get desired biophysical parameter, LAI. The daily LAI from INSAT CCD data was retrieved using radiative transfer simulation. By applying wheat mask we can segregate the wheat LAI from the agriculture LAI. From daily wheat LAI of Indo-gangetic plane, maximum wheat LAI of the season was estimated using the LAI from January to March (Figure 17) The retrieval of LST was rst carried out using NOAA AVHRR LAC (Local Area Coverage) data at 1km x 1km and validated over LASPEX sites in Gujarat by comparison with tower-based air and surface soil temperatures (Bhattacharya and Dadhwal, 2003). Later, mono-window method was used to retrieve LST from Kalpana-1 VHRR thermal data and compared with MODIS LST within 3C (Bhattacharya et al., 2009). Surface soil moisture in cropped soil was estimated using a soil wetness index (SWI) based on LST-NDVI 2D scatter at eld-scale (~90 m) and landscape-scale using ASTER and MODIS (~ 1km) products. The validation

regional-scale microwave-radiometer (C-band) soil moisture product from AMSR-E showed a good match (correlation coefcient = 0.75, RMSE = 0.027) for fractional vegetation cover upto 0.5 (Mallick et al., 2009). Satellite-based simplied evaporative fraction based surface energy balance was used to estimate clear-sky daily actual and relative evapo-transpiration over agricultural landscapes (~ 1km) and regional-scale (~ 8 km) agricultural land uses using MODIS (Mallick et al., 2009) and Kalpana-1 ground-to-MODIS-to-Kalpana produced an error of 0.8 mmd-1 VHRR (Bhattacharya et al., 2010) optical and thermal data, respectively. Two-step validation from

Forewarning of disease is the need of the hour. Weather based assessment of disease conduciveness was evaluated. The remote sensing of pests and diseases has remained a challenge due to the complexity of occurrence, overlapping with other factors, varying magnitudes and subtle manifestation. The mustard aphid (Lypaphis erysimi) infestation models have been developed from near-surface air temperature and relative humidity from sounder data (e.g. TOVS), and sowing dates (Bhattacharya et al., 2007) and validated through a collaborative study with National Research Centre on
Fig. 18: Regional mustard aphid onset over north-west India

Rapeseeds and Mustard (NRCRM), Bharatpur, Rajasthan. These models were later upscaled

and extrapolated using SPOT-VGT to map aphid onset dates (Dutta et al., 2008, Figure 18). A new methodology of multi-stage tracking of Sclerotinia rot (Sclerotinia sclerotiorum) disease in a large mustard growing region over Bharatpur district has been conceptualized and demonstrated that showed the spatial variability of conducive temperature and humidity level for pest infestation in wheat growing regions.

Crop Assessment: Beyond India

In-season rice area estimation using C-band Synthetic Aperture Radar (SAR) data is being done in India for more than a decade. Since the rice crop growing environment in India is a diverse one in the world having all the rice cultural types, the rice backscatter is quite exhaustive. A long term backscatter signature bank of rice elds has been developed. The well calibrated backscatter signature bank thus has the potential to act as seed for extending signature to classify rice lands of other Asian countries. The signature of the rice crop in Southern Bangladesh matched well with that of Gangetic West Bengal and the signature of rice crop in the Sub-Himalayan West Bengal particularly that of Dinajpur and Maldah districts matched well with that of the northern area of Bangladesh. A sample segment 71

N N R M S

Demonstration of Technique for Forewarning Pests and Diseases

B U L L E T I N

M A R C H

2 0 1 3

with in situ measurements showed errors of 0.039 and 0.033, respectively. The comparison with

approach was used to estimate the country level acreage. A 20 x 20 km grid was used and 10 % of grid was selected randomly (Figure 19). Acreage estimation was done using the standard random sampling aggregation using crop statistics of each selected segments. In total, 27 sample segments were analysed. The country level acreage estimated using statistical aggregation of random sampling resulted in acreage of 5.42 mha which is 5.0 per cent of the reported rice acreage for the same year. The acreage estimation of other crops in different parts of the world is also being explored.

Institutionalisation and Human Resources Development

The institutionalisation of remote sensing in agriculture was given deep thought right from the beginning. Initially educational, academia and research institutes were roped in to aid in research and feed to further development of programmes. The need for state remote sensing centres and regional
Fig. 19: Multidate SAR FCC overlaid with 20 X 20 km sample frame overlaid

remote sensing service centres was then visualised and brought into implementation.

Apart from this, state agricultural universities and agricultural departments were also involved in these projects in groundtruth collection and image analysis. These interactions led to human resource development in RS applications of agriculture. The periodic interaction with collaborating and user agencies and keeping them informed about their need and technological strength/advances resulted in formation of MNCFC, New Delhi and geoinformatics lab in NHRDF, Nashik.

Conclusions
Remote sensing applications of agriculture expanded into different domains and further many of them grew to higher levels of maturity during last twenty-ve years. The crop production forecasting for example started from experimental stage and moved upto operational stage. The 70s was a experimental phase while the 80s was formulation of RS applications projects. In the 90s these were tested on large area and operationalised. Programme like FASAL was transferred to MNCFC under Ministry of Agriculture. The newer applications of agriculture and its environment assessment were also explored. With availability of SAR sensors, monitoring of crop during kharif season became a reality. India is a global leader in agricultural applications of remote sensing and has carried out capacity building not only in India, but also in other countries. The specic tools and techniques have been developed to cater to above needs and operationalisation. Almost all sensors spanning the entire range of EMS used in RS application have been studied and host of them have been showcased. The emphasis in future should be on products and services sector encompassing more decipherable and ready to used knowledge based RS products. The agroecosystems analysis and climate change impacts would be the focal theme in which variety of components can come from RS data. The horticulture and site specic management would demand much

would fuse in the future to deliver the near-realtime service to all stakeholders of agriculture.

Authors are thankful to scientists of SAC and NRSC, ofcials of Ministry of Agriculture, scientists of collaborating institutes for valuable contributions to the project. The encouragement, technical guidance and support of Shri. A. S. Kiran Kumar, director, SAC and Dr. R. R. Navalgund, former director, SAC is highly acknowledged.

Agrawal, R., Jain, R.C. and Jha, M.P. (1983). Joint effects of weather variables on rice yields. Mausam, vol. 34, pp 189-194. Bhattacharya, B. K. and Dadhwal, V. K. (2003). Retrieval and validation of land surface temperature (LST) from NOAA AVHRR thermal images of Gujarat, India. International Journal of Remote Sensing. 24(6), 1197-1206 Bhattacharya, B. K., Dutta, S., Dadhwal, V.K., Parihar, J.S., Chattopadhyay, C., Agrawal, V., Kumar, V., Khan, S.A., Roy, S., Shekhar, C. (2007). Predicting Aphid (Lipaphis erysimi) Growth in Oilseed Brassica Using Near Surface Meteorological Data from NOAA TOVS A Case Study. International Journal of Remote Sensing, 28(17-18), 3759 3773 Bhattacharya, B.K., Mallick, K., Padmanavan, N., Patel, N.K., and Parihar, J.S. (2009). Simple retrieval of land surface albedo and temperature using data from Indian geostationary satellite - a case study for winter months. International Journal of Remote Sensing, 30(12), 3239-3257 Bhattacharya, B.K., Mallick, S., Patel, N.K. and Parihar,J.S. (2010). Regional evapotranspiration over agricultural land using remote sensing data from Indian geostationary meteorological satellite. Journal of Hydrology, 387, 65-80 Chaurasia, S., Bhattacharya, B.K., Dadhwal, V.K. and Parihar, J.S. (2006). Field-scale leaf area index estimation using IRS 1D LISS III data. International Journal of Remote Sensing, vol. 27, pp 637-644. Chaurasia, S., Nigam, R., Bhattacharya, B.K., Sridhar, V.N., Mallick, K., Vyas, S.P., Patel, N.K., Mukherjee, J., Shekhar, C., Kumar, D., Singh, K.R.P., Bairagi, G.D., Purohit, N.L. and Parihar, J.S. (2011). Development of regional wheat VI-LAI models using Resourcesat-1 data. Journal of Earth System Science 120 (6), 1113-1125 Chhabra, A., Manjunath K.R. and Panigrahy, S. (2009). Assessing the role of Indian livestock in climate change. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, (eds. Sushma Panigrahy, Shibendu Shankar Ray and Jai Singh Parihar), Vol. XXXVIII Part 8/ W3, pp. 359-365. Chhabra, A., Manjunath, K.R., Panigrahy, S. and Parihar, J.S. (2009). Spatial pattern of methane missions from Indian livestock. Current Science, 96(5): 683-689. Chhabra, A., Manjunath, K.R., Panigrahy, S. (2010). Non-point source pollution in Indian agriculture: Estimation of nitrogen losses from rice crop using remote sensing and GIS. International Journal of Applied Earth Observation and Geoinformation 12: 190200. Dutta, S., Bhattacharya, B. K., Rajak, D.R., Chattopadhyay, C., Dadhwal, V.K., Patel, N.K., Parihar, J.S. and Verma, R.S. (2008). Modelling regional level spatial distribution of aphid (Lipaphis erysimi) growth in Indian mustard using satellite-based remote sensing data. International Journal of Pest Management, 54(1), 51 62

73

N N R M S

B U L L E T I N

References

M A R C H

Acknowledgements

2 0 1 3

more complex algorithms and service oriented products. The RS and communications technologies

Mallick K., Bhattacharya, B.K., Rao, V.U.M., Reddy, D.R, Banerjee, S., Hoshali, V., Pandey, V., Kar G., Mukherjee, J., Vyas, S.P., Gadgil, A. S. and Patel, N.K. (2009). Latent heat ux estimation in clear sky days over Indian agroecosystems using noontime satellite remote sensing data. Agricultural and Forest Meteorology, 149, 1646 -1665 Mallick, K., Bhattacharya, B.K. and Patel, N.K. (2009a). Estimating volumetric surface moisture content in cropped soils using a soil wetness index based on surface temperature and NDVI. Agricultural and Forest Meteorology,137, 1327-1342. Manjunath K. R., Panigrahy S., Kundan Kumari, Adhya T.K. and Parihar J.S., (2006). Spatio-temporal Modelling of Methane Flux from the Rice Fields in India using Remote Sensing and GIS. International Journal of Remote Sensing. 27 (20): 4701-4707. Manjunath, K. R., Panigrahy, S., Adhya, T.K., Beri, V., Rao K. V. and Parihar, J.S. (2011). Methane Emission Pattern of Indian Rice-Ecosystems. Journal of Indian Society of Remote Sensing: 39(3), 307-313. Manjunath, K.R. and Panigrahy, S. (2009). Spatial database generation of the rice-cropping pattern of India using satellite remote sensing data. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, (eds. Sushma Panigrahy, Shibendu Shankar Ray and Jai Singh Parihar), Vol. XXXVIII Part 8/W3, pp. 262-268. Manjunath, K.R., Potdar, M.B and Purohit, N.L, (2002). Large Area Operational Wheat Yield Model Development and Validation based on Spectral and Meteorological Data, International Journal of Remote Sensing, Vol. 23, No. 15: 3023-3038. Navalgund, R.R., Parihar J.S., Ajai and Nageshwara Rao, P.P. (1991). Crop inventory using remotely sensed data. Current Science, vol. 61, pp 162-171. Oza, M.P., Bhagia, N., Rajak, D.R. and Dadhwal, V.K. (2002). All India wheat inventory using multidate IRS WiFS and weather data. Proceedings of the ISPRS Comm. VII Symposium Resource and Environmental Monitoring held at Hyderabad, India during Dec. 3-6 2002, IAPRS & GIS, vol. 34, Part 7, pp 338-343. Panigrahy, S., Manjunath, K R, Singh, R. P., Chhabra, Abha and Parihar, J S. (2010a). Spatio-temporal pattern of Green House Gases over India and upscaling of methane emission from agriculture using space technology. Earth observation Applications in the Climate Change Studies, Bulletin of the National Natural Resources Management System, NNRMS (B) -35, Department of Space, Bangalore, India, December 2010, pp-50-56 Panigrahy, S., Upadhyay, Gargi, Ray, S. S. and Parihar, J. S., (2010). Mapping of cropping system for the Indo-gangetic plain using multi-date SPOT NDVI-VGT Data. Journal of Indian Society of Remote Sensing. 38, pp. 627-632. Panigrahy, S., Ray, S. S., Manjunath, K. R., Pandey, P.S., Sharma, S.K., Sood, A., Yadav, M., Gupta, P.C., Kundu, N. and Parihar, J.S. (2009). A spatial database of cropping system and its characteristics to aid to climate change impact assessment studies. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, (eds. Sushma Panigrahy, Shibendu Shankar Ray and Jai Singh Parihar), Vol. XXXVIII Part 8/W3, pp. 150-155. Parihar, J. S. and Dadhwal, V. K. (2002). Crop production forecasting using remote sensing data: Indian experience. Proceedings of the ISPRS Comm. VII Symposium Resource and Environmental Monitoring held at Hyderabad, India during Dec. 3-6 2002, IAPRS & GIS, vol. 34, Part 7, pp. 1-6. Patel, N.K., Chakraborty, M., Dutta, S., Patnaik, C., Parihar, J.S., Moharana, S.C., Das, A., Sarangi, B.K. and Behra, G. (2004). Multiple forecasts of kharif rice in Orissa state Four year experience of FASAL pilot study, Journal of the Indian Society of Remote Sensing, vol. 32, pp 125-143. Sahai, B. Chandrsekhar, S., Barde, N.K, and nagBhusana, S.R. (1977). Agricultural resources inventory and surveys experiment. In: Rycroft M. J. and Stickland A. C. (eds) COSPAR Space Resarch, pergaman press, Oxford, pp: 3-8. Sahai, B. and Dadhwal, V. K. (1990). Remote sensing in Agriculture. In Technology blending and agrarian prosperiety (Eds Verma J. P. and Verma A.), Malhotra Publication, New Delhi, pp. 83-98.

of rabi crop progress using geostationary satellite INSAT 3A CCD data. SAC, Scientic note: ABHG/ SAC/SN/03/2011. Panigrahy, S., and Manjunath, K. R. (2009). Horticulture studies (2000-2008), Highlights of the results and achievements of the project. EOAM/SAC/AMD/Hort/PR/01/2009. Tripathy, R., Ray, S. S., Kaur, H., Jalota, S. K., Bal, S. K. and Panigrahy, S. (2011). Understanding spatial variability of cropping system response to climate change in Punjab state of India using remote sensing and Terrestrial Ecosystems (eds Panigrahy, S., Ray, S. S. and Huete, A. R.), pp. 2933. Tripathy, Rojalin, Chaudhari, Karshan N., Mukherjee, Joydeep, Ray, Shibendu S. Patel, N. K., Panigrahy, Sushma & Parihar, Jai Singh (2013): Forecasting wheat yield in Punjab state of India by combining crop simulation model WOFOST and remotely sensed inputs, Remote Sensing Letters, 4:1, 19-28. data and simulation model. In ISPRS Archives XXXVIII-8/W20, Workshop Proceedings: Earth Observation

75

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Vyas Swapnil, Rahul Nigam, Patel, N. K., Sushma Panigrahy and Jai Singh Parihar (2011). Monitoring

LAND RESOURCES MANAGEMENT AND MONITORING

Ravisankar T, Sreenivas K, Ravishankar G, Fyzee MA, Sujatha G, Rajiv Kumar, Manoj Raj Saxena, Shyam Sunder B, Padma Rani G National Remote Sensing Centre / ISRO Balanagar, Hyderabad -500 037, India Email: ravisankar_t@nrsc.gov.in

Introduction
Natural resources inventory is basic information required for any resource planning. During the past three decades several efforts were made to employ remote sensing data for inventory of land use / cover, soils and assessing other land related problems in India. An attempt has been made in this article to assess the utilization of remotely sensed data from Indian Remote Sensing Satellite (IRS) sensors since late eighties along with the improvements in methodologies that took place in themes like land use land cover, soils, land degradation, wastelands and watershed management.

Land use Land Cover (LULC) Mapping

Land use / cover is one of the major inputs to any resource planning and very dynamic both seasonally as well as annually. Realizing the need for an up-to-date nationwide LULC maps by several departments in the country, a LULC classication system with 24 categories up to Level-II, suitable for mapping on 1:250K scale, was developed by NRSC in consultation with various user departments (Ravishankar et al., 2003). The district-wise LULC analysis was completed using satellite data for the period 1988 89 for the Planning Commission of India. Besides, LULC mapping on 1:50K scale has been carried out in Integrated Mission for Sustainable Development (IMSD) project covering about 25% area of the country (NRSC,2002a) and in 17 states under NRIS project (continuation to IMSD) and Integrated Resources Information System for Desert areas (IRIS-DA) covering parts of four states Rajasthan, Karnataka, Gujarat and Haryana. Subsequently, before taking up of national level NR Census Mission, the proto type districts studies were taken up to develop legends for mapping various themes, standardization of methodology including digital data base creation and to generate census statistics for various natural resources. Rapid Assessment of Land Use / Land Cover on 1: 250K scale was initiated annually in 2004-05 time frame using multi-temporal IRS AWiFS data for entire country. In this project the cropped areas during kharif, rabi and zaid seasons are assessed along with area under double/ triple crops and at the end of the year net sown area is estimated. So far eight cycles have been completed and the data is made available to users through Bhuvan. A composite Land Use/ Land Cover data sets has been generated using multi temporal satellite data (NRSC,2012). Analysis are being carried out on 1: 250K scale using level I and some Level II classes of land use classication. Figure 1 shows the integrated land use land cover map of India for the year 2011-12.

Census (NRC) project on LULC map of India on 1:50K scale addressing 9 classes at Level 1, 35 classes at Level 2 and 79 classes of land use at Level 3 was carried out. It involved interpretation of about 850 Resourcesat-1 LISS-III satellite image frames covering Kharif, Rabi and Zaid seasons of referencing of three season Resourcesat-1 LISS-III images, On-screen preliminary visual interpretation of LISS-III images, ground truth collection, map nalization, in-process quality
Fig. 1: Land Use Land cover map of India of 2011-12 generated from IRSAWiFS data

assurance mechanism, database organisation and spatial statistics generation. LULC classes were evolved and dened through a series of

technical deliberations involving national and state level users and R & D institutions. An atlas was generated after completion of the project which presents the national, state and district level estimates of area under each LULC category for the 29 States and 6 Union Territories for the period of 2005 06. Figure 2 shows the land use land cover map generated at 1:50K scale for the whole country. Area under agriculture in the country is about 18,00,000 sq.km. and the long fallow is very negligible, which is as low as 60 sq.km. for the whole country (NRSC, 2011a). In the present exercise, an area of approximately 1,21,000 sq.km. in Jammu & Kashmir could not be mapped, which is roughly 3.7 per cent of the total geographical area of the country The mapping will be repeated under the Natural Resources Census for the year 2010-11 in the next cycle. Under Space based Information System for Decentralized Planning (SIS_ DP) project, LULC information on 1:10K scale is being generated using IRS-LISS-IV Resourcesat-1/2 and Cartosat-1 data where broad LULC classes amenable to Panchayatraj level planning.

Soil Resources
For efcient management, we must have an in-depth knowledge about different soils, their morphology, physical and chemical
Fig. 2: Land Use Land Cover Map of India generated from IRS-LISS-III three seasons data.

properties, behaviour, kind and degree of problem and their extent and distribution on

77

N N R M S

B U L L E T I N

2005 06. The approach involved Geo-

M A R C H

2 0 1 3

Under NRR Natural Resources

the landscape, which can be achieved through soil survey and mapping. The role remote sensing was proved to be a powerful tool to study resources in spatial domain in time and cost effective manner. In India, initial attempts were made with aerial photos to map soil resources. Satellite data from IRS sensors are being used to generate soil maps through monoscopic (non-stereoscopic) visual interpretation and computer-assisted digital analysis approaches. In visual interpretation approach, the intimate relationship between physiography (landform) of the terrain and soils occurring therein is exploited. The digital analysis is based on the premise that the each object (here individual soil unit) could be characterized by its unique spectral response pattern. Digital approach was used to generate soil resources maps and to delineate problem soils. The launch of the Indian Remote Sensing Satellite (IRS-1A, 1B, 1C, 1D) with sensors, namely Linear Imaging and Self-scanning Sensors (LISS-I, II, III, IV) and Panchromatic (PAN) has opened new vistas in soil and land degradation mapping in a rapid and cost-effective way at different scales from 1:250K to 1:10K (Karale,1992; Rao and Venkataratnam, 1992). The multispectral data acquired during summer season and peak crop growing periods are ideal for soil resources mapping, since land cover and its vigor explains the inherent variability of soils within broad physiographic units. Regional-Level Soil Mapping During eighties, NRSA had prepared soil maps for various parts of the country like part of Orissa, Andhra Pradesh, Karnataka, Haryana and north eastern states of India at 1:250K scale using satellite data. National Bureau of Soil Survey and Land Use Planning (NBSS&LUP) in cooperation with the Soil Survey staff from different states, completed a massive programme of mapping soils on 1:250K in India. The regional soil mapping comprises image interpretation (IRS / Landsat) and landform analysis, eld surveys, laboratory analysis followed by soil correlation at soil series / sub-group level to prepare soil map showing their association. These soil maps are very useful for planning at regional level and for identifying regions of potentials and limitations. Sub-regional Soil Mapping During the period from mid eighties to mid nineties, the second generation of remote sensing satellites viz., IRS satellites were launched with higher spatial and spectral resolutions which had enabled to map soils at 1:50K scale to the level of association of soil series. The studies with IRS satellites -1A /1B LISS-II had set the trends of rapid development and wider acceptability of remote sensing application in soil resources study. At National Remote Sensing Centre, soil maps were prepared at 1:50K scales on operational basis in various parts of the country for various user departments like Agriculture, Command Area Authorities, etc. The soil mapping was carried for specic purposes like land capability classication, land irrigability assessment, optimum land use planning, etc. The agship project for sub-regional / district level planning using LISS-II data was the Integrated Mission for Sustainable Development (IMSD) Project of Department of Space, where the soil resources were mapped at 1:50K scale for various states in India like Andhra Pradesh, Karnataka, Tamil Nadu, Uttar Pradesh, Madhya Pradesh, Haryana etc. covering an area of approximately 83.3 mha (DOS, 1998). Post soil map preparation, the soils were evaluated for land capability classication, land irrigability assessment and suitability to major crops under IMSD project. This soil information was used as key input in the generation of action plans for sustainable development of land and water resources. Besides, the LISS-II data was extensively used for soil mapping in command area projects. Before introduction of irrigation in any proposed command areas, soil survey is specically carried out, to evaluate the soils of the proposed command area to study their suitability for irrigation. This enables identication areas having potential for secondary salinization & preventing salinity or water logging development so as to maximize the crop production by efcient utilization of irrigation water. NRSA (1997a, 1997b) had carried out land irrigability assessment study in Sri Ram Sagar Project (SRSP) Phase-II and Krishna-Pennar link canal command area in Andhra Pradesh in India using remotely sensed

and assessed for their suitability to crops like jowar, chillies, maize, groundnut, cotton, pulses, paddy etc., by following the FAO approach. The land suitability for irrigation was also assessed. Similar studies were carried out using LISS-II and LISS-III data for soil mapping and problem area identication in tribal areas as well as for its development planning at 1:50K scale. Semi-detailed Soil Mapping From mid nineties onwards with the availability of 5.8 m (PAN Sensor) and 23.5 m spatial prepare soil maps at 1:12.5K (NRSA, 1997c) and 1:25K scales (NRSA, 2002b) with PAN merged LISSIII data. In a few such studies, soils were mapped at 1:12.5K scale for a part of Sanga Reddy and Kurnool districts, Andhra Pradesh, Dadra and Nagar Haveli Union Territory (UT) and under National Agricultural Technological Project (NATP) for different micro-watersheds. At this scale, the soils could be mapped at the level of soil series with association of soil phases. The soil information was used in the perspective land use planning and planning crop productivity improvement. Under National Urban Information System (NUIS) project soil maps were generated depicting soil type, depth and textural information useful for broad land capability assessment around urban areas on 1:10K scale using IRS-LISS-IV and Cartosat-1 data. The capability of LISS-IV + Cartosat-1 for preparing soil maps at 1:10K scale showing soil phases required for managing soils at Village level along with overlaying of cadastral maps was demonstrated under Village Resource Centre (VRC) project. An attempt has been made to understand the potential of high resolution satellite data (LISS-IV) to depict the spatial variability of soil fertility. Further, the village cadasters were overlaid on soil / action plan maps to draw cadastre-wise information. The above studies indicate that IRS data from LISS-II and LISS-III was extensively used for soil mapping at 1:50K scale for major part of the country. However, there is strong requirement for village-level resource planning which calls for mapping at 1:5K to 1:10K scales. This effort requires huge infrastructure and scientic map power and can be enabled with LISS-IV and PAN merged data and satellite data with 1 m spatial resolution. resolution data (LISS-III sensor) from IRS-1C/1D satellites, experiments have been carried out to

Watershed Management
Watershed management has been dened as the integrated use of land, vegetation, and water in a geographically discrete drainage area for the benet of the people, with the objective of protecting or conserving the hydrologic services which the watershed provides and of reducing or avoiding negative downstream or groundwater impacts (World Bank, 2007). It addresses key issues related to land and water resources and incorporation of environmentally benign or clean technologies viz. organic agriculture processes and agricultural activities, livestock, forestry, industrial and agroindustrial. It consists of selection of watershed, watershed characterization, watershed prioritization, analysis of natural resources, identication of constraints, development of action plan, implementation of action plan, monitoring, and impact assessment. LISS-I, II,III,IV and Cartosat-1 from Indian Remote Sensing Satellites have been successfully employed over the last 25 year ( NRSA, 1996; Saxena et al., 2000, NRSA, 2006; Thakkar and Dhiman, 2007) for the above activities. Watersheds are traditionally delineated manually based on contour information and drainage depicted in topographic maps. Recently with the availability of digital topographic maps, efforts are going on to generate automatic delineation of watersheds due to technological advancements. The 79

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

data from IRS-LISS-II / III sensors data at 1:50K scale. The soils of the command areas were mapped

DEM generated from stereo pairs Cartosat-1 mission with 2.5 m horizontal resolution stereo images are some of the examples of such a capability. Watershed prioritization is, basically, the ranking of different sub-watersheds of a watershed according to the order in which they have to be taken up for treatment and /or soil and water conservation measures with limited nancial resources available. Watershed characterization involves inventorying and assessment of natural resources viz., soils, land cover / land use, water, crops etc. Department of Space had demonstrated the application of geospatial technologies viz., remote sensing, GPS and GIS in watershed management through projects like Sujala watershed project in Karnataka, National Agricultural Technology Project (NATP) and Integrated Mission for Sustainable Development (IMSD) etc. (NRSA, 2006; Roy et al., 2010; Dwivedi and Ravisankar, 2010) Sujala Watershed Project in Karnataka: SUJALA is a community driven, participatory watershed development project, implemented in 854 micro watersheds distributed across 38 taluks in ve districts of Karnataka. The monitoring system adopted in Sujala project is unique and has a blend of modern technologies like Remote Sensing (RS), Geographical Information System (GIS), and Management Information System (MIS) with conventional ground based monitoring system. The monitoring system in SUJALA has three components viz., input output monitoring, process monitoring and impact assessment. Sukriya- a bilingual software package has been developed for action plan preparation where in systematic beneciary database is created for implementation, monitoring and impact assessment. Impact is evaluated using a set of qualitative and quantitative indicators before, during and after the project implementation using remote sensing and GIS technology. The integrated approach of monitoring and evaluation adopted in Sujala has provided the state-of-the-art information enhancements for tracking the project impacts and outcomes to answer questions about progress against broad development indicators. National Agricultural Technology Project (NATP) Crop Production System Approach in Watershed Development: This production systems-based approach is the special feature that distinguishes watershed development from earlier plot/ eld-based approaches to soil and water management. It aims to optimize moisture retention and reduce soil erosion, thus maximizing productivity and minimizing land degradation. Under NATP programme, the problems / potentials of the rainfed rice, cotton, pulses, oilseeds and kharif and rabi sorghum crop growing areas are studied using remote sensing and GIS techniques on watershed basis at 1:50K scale and micro-watershed at 1:12.5K scale under each watershed (NRSC, 2006). The work was carried out in 11 watersheds covering total geographical area of 13.5 lakh hectares in the states of Andhra Pradesh, Maharastra, Madhya Pradesh, Orissa, West Bengal, Jharkhand, Chhattisgarh, Gujarat and Assam. Under NATP project, in all watersheds / micro watersheds natural resources such as soils, land use / land cover, hydro-geomorphology and drainage were mapped at 1: 50K scale (IRS-LISS data) / 1:12.5K scale (IRS- PAN + LISS-III data) using satellite remote sensing data, supported by ground truth and ancillary data. Action plans were developed for watershed and micro-watershed broadly covered various aspects of crops, soil and water conservation. Integrated Watershed Management Project (IWMP): Ministry of Rural Development (MRD), Govt., of India recently launched this project by pooling all watershed programmes under Desert Prone Area Programme (DPAP), Desert Development Programme (DDP) etc. This project strongly advocates using of remote sensing and GIS technologies from preparation of Detailed Project Reports (DPRs), characterization of resources, development of soil and water conservation plans and impact assessment. A xed percentage of funds allocated under IWMP scheme to different states has to be spent on use of RS and GIS tools in watershed development. As a part of the project capacity building programmes were also taken up by Dept., of Land Resources, MRD for training Govt., ofcials in the use of RS and GIS tools. Recently with the availability of SIS_DP data products on 1:10K scale (LISS-IV+Cartosat-1 merged products), the role of remote sensing technology in watershed projects is greatly enhanced.

resolution satellite data could be effectively used for watershed management and monitoring activities at land ownership level. Further such data could be of immense help in tracking the implementation, apply midcourse corrections and for assessing long-term effectiveness of the programme implemented. The synergy of GIS and Web Technology allows access to dynamic geospatial watershed information without burdening the users with complicated and expensive software. Advancements in various technological, scientic, modelling, communication and data dissemination fronts should be integrated on to a seamless platform to develop an Intelligent Watershed Management System.

Land Degradation
Land degradation has been dened by UNEP (1992) as a human induced or natural process that negatively affects the land to function effectively. It is the temporary or permanent lowering of the productive capacity of land. The land degradation processes that are amenable to remote sensing techniques are water erosion, wind erosion, waterlogging, salinization/alkalization, acidication, frost heaving / frost shattering, anthropogenic, mass movement, etc. Salt-Affected Soils Several efforts were made during 80s to understand the potential multi-spectral satellite data to delineate salt-affected soils. These initial efforts to delineate salt-affected soils using remote sensing data culminated to a national project on Mapping of salt-affected soils of India, where in salt-affected soils and their type and magnitude were mapped at 1:250K scale for entire India. Subsequently, efforts were also made to monitor salt affected soils in spatial domain using multi temporal data (Dwivedi and Venkataratnam, 1992; NRSA, 2000; NRSA, 2001). Realizing the potential of remote sensing data in mapping salt-effected soils and water logging, studies were carried out in (a) south coastal districts of Andhra Pradesh viz., Krishna, Guntur, Prakasam and Nellore (b) command areas of Sharda Sahayak, Chambal, Jayakwadi, Purna, Upper Tapi, Bhima, Krishna, NSP, and Girna etc., in various states at 1:50K scale using IRS LISS-II/III data; (c) NR Census prototype studies in Guntur district of Andhra Pradesh Large scale mapping of salt-affected soils was attempted in part of Uttar Pradesh (India) using IRS-1C/1D LISS-III and PAN-merged data in collaboration with the Remote Sensing Applications Centre, U.P. Besides, the IRS LISS-III + PAN merged data was successfully used for monitoring the success and progress of the ongoing land reclamation programme under a World Bank-aided project. Soil Erosion Satellite data has also been utilized in qualitative (NRSA, 1990) and quantitative (Sreenivas, 2001) assessment of soil erosion through spatial modelling. Delineation of various intensities of erosion namely, rill eroded areas, gullied lands as well as ravines were successfully attempted using multispectral IRS data through visual interpretation. Cartosat-1 stereo was used of improve the delineation of ravines as well as for stereoscopic measurements of their bed width and depth. Similarly, IRS LISS- III data derived vegetation cover and soil information along with DEM was used to assess the impact of mining on forest and adjacent crop lands using Erosion-Deposition model. IRS LISS-III and Landsat-TM data acquired at and interval of 10 years was successfully used to monitor the treated watersheds through digital techniques.

81

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

The remarkable developments in space technology currently with the high spatial and temporal

Shifting Cultivation Studies on mapping and monitoring of shifting cultivation areas using multispectral data have been carried since 1976 using both visual and digital techniques (Dwivedi and Ravisankar, 1991; Dwivedi et al., 1992) demonstrated the potential of remotely sensed data in the study of stages of shifting cultivation and role of image scale on delineation of jhum lands and monitoring of jhum lands in parts of Tripura state. Waterlogging Satellite data has successfully used to map and monitor waterlogged areas. However, space borne multispectral data in visible and NIR bands was found to be of limited use in black soil areas because of the similarities in spectral response of non-waterlogged black soil areas. Monitoring of waterlogged areas under major commands had been carried out using multi-temporal LISS-II / III data (NRSA, 2000). The above efforts and experiences have culminated to a national mapping project on land degradation, where multi-temporal LISS-III data has been used to map land degradation at 1:50K scale under NR Census programme. About 8 land degradation processes with 32 land degradation types along with severity classes. Out of various land degradation types mapped, water erosion is the single dominant process operating India.

Wastelands Mapping
National Wastelands Development Board dened waste land as degraded land that can be brought under vegetative cover with reasonable effort and which is currently under-utilized and land which is deteriorating due to lack of appropriate water and soil management or on account of natural causes. In a given set of geographical conditions, it is essential to understand the various pathways or the processes of land degradation that relegate land areas as wastelands and their relevance to the socio economics of human society. Consequently, development of wastelands can become directly relevant to sustainable development of natural resources. At the instance of Ministry of Rural Development (MRD), Government of India, NRSC had mapped wastelands at 1:50K scale in different years and has brought out the National Wastelands Atlases during 2000, 2005, 2010 and 2011, depicting the spatial distribution of wastelands in the country for the periods 1986 2000, 2003, 2005 2006 and 2008-09, respectively. National Remote Sensing Agency, Department of Space, has prepared the wastelands maps of all the states and Union Territories on 1:1 Million scale in the year 1985 using 1980-82 data sets. About 190 single season (rabi) Landsat satellite images (MSS) acquired during 1980-82 were used in the study. An eight fold classication system was adopted for mapping of wastelands in the country. Based on this study, the total area under wastelands in the country was estimated to be 53.3 mha or 16.20 per cent of the total geographical area of the country. These wastelands maps due to small scale (1:1 million) and low resolution of the data used (80 m), any isolated patch of land less than 100 ha could not be mapped. The maps were of limited use for reclamation at micro level. Mapping of wastelands on 1:50K scale was undertaken in ve phases which was spread over 1986 to 2000. The 1 : 1 million scale wastelands maps prepared by NRSA helped to prioritize districts in the country for mapping at 1:50K scale . The wastelands information so generated in ve phases using single season satellite data were consolidated and presented in the form of an atlas in May, 2000. A nal consolidated atlas on wastelands was published in May, 2000 where wastelands were reported as 63.85 mha constituting 20.17 % of total geographical area of the country. Subsequently, National Wastelands Updation Mission was initiated in 2003 to monitor the changes in wastelands at the request of MRD. Single season Indian Remote Sensing LISS III satellite data acquired during Rabi season of 2003 has been used to map the wastelands for the entire country. The task force constituted for the purpose recommended a

Central Arid Zone Research Institute (CAZRI). The extent of wastelands in the country was estimated as 55.27 mha. (NRSA and Ministry of Rural Development, 2005). To understand the spatial and temporal changes in wastelands, a project on National Wastelands Monitoring was initiated in 2006 with the objective of assessing the status of wastelands and to monitor its changes. For the rst time, three seasons IRS LISS III satellite data viz kharif, rabi and zaid of 2005-06 has been used to derive information on wastelands. This exercise enabled improvement due to use of three season satellite data and improvement in interpretation techniques. Also, this mapping exercise was carried out totally in GIS domain adopting Open Series GIS specication to enable wider use /dissemination of information. An area of 47.22 mha (14.91% of TGA) was recorded as wastelands (NRSC, 2010). In the year 2010 wasteland mapping was carried out using IRS LISS-III data of 200809 (Figure 3) and to bring out changes in wastelands with reference to 2005-06. This process involved the overlay of Wasteland GIS layer of 2005-06 on the multi season satellite data of 2008-09 and to record the changes using change vector approach. A uniform mapping procedures and standards have been adopted in the study. The wasteland classes of 2005-06 were updated using satellite data of
Fig. 3: Wasteland Map of India of 2008-09 generated from IRS-LISS data

2008-09 and the change areas were identied. These changes were later conrmed using

limited ground checks. About 32,000 sq. km of wasteland has undergone a positive change in the entire country during the period, by getting converted into non-wasteland. The declining trend was observed in 21 states of the country (NRSC, 2011b). Majority of the change areas have been found in the states of Rajasthan (10264 sq.km), Mizoram (2669 sq.km.), Manipur (2391 sq.km.), Gujarat (2858 sq.km.), Bihar (1895 sq.km.), Karnataka (1478 sq.km.), Andhra Pradesh (1682 sq.km.) and Uttar Pradesh (1269 sq.km.). Bulk of area (20,000 sq. km.) from different wasteland categories has changed into cropland (including fallow) class. A considerable area change in case of degraded forest-scrub dominant has been noticed getting converted into forest-dense/open & forest plantation classes (9,600 sq. km.) during the period of study. About 800 sq. km. of wasteland has been converted into other plantations. However, at the same time, a slight increase in total wasteland area was recorded in 8 states, where non-wasteland has been observed to have changed into wasteland, bringing down the net decrease in wastelands during 2008-09 to about 0.17 percent, compared to 2005-06.

Conclusion
Spatial technologies in land resources assessment has reached operational stage. All the sensors from IRS satellite viz., LISS-I/II/III/IV, AWiFS/WiFS and from Cartosat-1/2 have been exploited for mapping and monitoring of various resources and many national projects have been completed successfully 83

N N R M S

B U L L E T I N

in the delineation of wasteland categories compared to earlier exercises as there was improvement

M A R C H

2 0 1 3

28 fold classication for adopting in the exercise based on the pilot studies carried out by NRSA and

either on 1:250K scale or 1:50K scale. The data so generated are made available to users through Bhuvan services. The need for data bases on 1:10K scale may go up. The future efforts are directed in making use of microwave data and hyperspectral data for quantitative estimation of resources. The emphasis is on for automatic or semi-automatic feature extraction methods for land cover and modeling of land degradation processes in light of climatic change.

References
DOS, (1998). Application of Remote Sensing in soil resource mapping. Dwivedi, R.S. and Ravisankar, T. (1991). Monitoring shifting cultivation using space borne multispectral and multitemporal data. Int. J. Remote Sensing, 12 (3), 427-433. Dwivedi, R.S. and Venkataratnam, L. (1992). Land degradation studies using IRS-1A data. In R.L. Karale (ed.) Natural Resources Management; A New Perspective. Natural Resources Management System (NNRMS), Department of Space, Government of India, Bangalore. pp 153-169. Dwivedi, R.S. and Ravisankar, T. (2010). Watershed Management A Geospatial Technology Perspective. Chapter-3 in book titled Watershes Management, Restoratuin and Environmental Impact. Edit by Jermy C. Vaughn. Nova Science Publishers, Inc. New York. 2010. Karale, R.L. (1992). Remote sensing with IRS-1A in soil studies: Developments, status and prospects. In R.L. Karale(ed.) . Natural Resources Management - A New Perspective. National Natural Resources Management System (NNRMS),Department of Space, Government of India, Bangalore. pp128-143. National Remote Sensing Agency. (1996). Evaluation study of treated sub watersheds in Kundah, Lower Bhawani and Tungabhadra catchments using remote sensing techniques. Project report. National Remote Sensing Agency, Department of Space, Government of India. National Remote Sensing Agency. (2006). Atlas National Agricultural Technology Project - Development of Regional Scale Watershed Plans and Methodologies for Identication of Critical areas for Prioritized Land Treatment in the Watersheds of Rainfed rice, Oilseeds, Pulses, Cotton and NCRL Production Systems. NRSA (1990). : IRS Utilisation programme ; soil erosion mapping. Project report. NRSA, (1995). Soil Resource mapping of part of Kurnool district, Andhra Pradesh IMSD project). NRSA, (1996a). Satellite assessment of waterlogging and soil salininty/alkalinity in Nagarjunasagar right canal command area. NRSA, (1996b). Salt affected soils in South coastal districts of Andhra Pradesh NRSA, (1996c). Soil survey and land evaluation for agricultural land use planning in tribal areas of Andhra Pradesh. NRSA, (1997a). Soil survey for Land irrigability assessment Sriram Sagar Project, Phase II, A.P. NRSA, (1997b). Thematic mapping of proposed Krishna Pennar link canal command area. NRSA, (1997c). Evaluation of IRS-1C data for mapping soil resources and degraded lands. NRSA, (2000). Mapping slat affected soils and waterlogged areas in parts of NSP Right canal command and Krishna western Delta. NRSA,(2001). Management of salt affected soils and rational land use at village level using remote sensing and GIS NRSA,(2002b). Perspective land use planning Dadra and Nagar Haveli UT (A Remote sensing based approach). NRSA (2002a). : Integrated Mission for Sustainable Development Path to Progress.

Space, Govt. of India, Hyderabad. NRSC (2010). : Wastelands Atlas of India. NRSC (2011a). Land Use Land Cover Atlas of India (Based on Multitemporal Satellite data of 2005-06). NRSC (2011b). Wastelands Atlas of India (change analysis based on temporal satellite data of 200506 and 2008-09). NRSC (2012). National Resources Census: National Land Use and Land Cover mapping using multi temporal AWIFS data Rao, B.R.M. and L. Venkataratnam (1992). Utility of IRS- 1A data in soil resources mapping. In R.L. Karale (ed.) Natural Resources Management-A New Perspective. National Natural Resources Management System (NNRMS), Department of Space, Government of India, Bangalore. pp 424-430. Ravi Shankar, G., Raghavaswamy, V. Nagaraja, R., Gautam, N.C., (2003) : Remote sensing and GIS for land use/land cover data generation at national level: An Indian experience. Land use/ land cover management in India. B.S. Publications, Hyderabad, India. Roy, P.S., Ravisankar,T. and Sreenivas, K. (2010). Advances in geospatial technologies in integrated watershed management. Proceedings of the seminar on Advances in watershed management at New Delhi., Feb., 2010. Saxena, R.K., Verma, K.S., Chary, G.R., Srivastava, R. and Barthwal. A.K. (2000). IRS-1C data application in watershed characterization and management. Int. J Remote Sensing, 21(17): 3197-3208. Sreenivas, K. (2000). Soil erosion modeling using remote sensing and geographic information system. Thesis submitted to JNTU for the award of PhD Thakkar, A.K. and Dhiman, S. D. (2007). Morphometric analysis and prioritization of miniwatersheds in Mohr watershed, Gujarat using remote sensing and GIS techniques. Journal of the Indian Society of Remote Sensing. 35 (4). UNEP (1992). United Nations Environment Programme. Rio declaration on environment and development. Made at the United Nations Conference on environment and development, Rio de Janeiro, Brazil, available on the world wide web. World Bank. (2007). Watershed Management Approaches, Policies, and Operations: Lessons for Scaling Up. A report by the Energy, Transport and Water Department, World Bank, Washington, DC.

85

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

NRSA and Ministry of Rural Development (2005). : Wastelands Atlas of India, 2005. NRSA, Dept. of

WAtER RESOURCES MANAGEMENt : EvOLUtION OF SPACE TECHNOLOGY APPLICAtIONS

Venkateshwar Rao V, Sharma JR and Raghavswamy V National Remote Sensing Centre (ISRO), Hyderabad - 500 037, India Email: vvrao@nrsc.gov.in

Introduction
Water is basic to man and his environment and it regenerates and redistributes through evaporation and rainfall. Water and the natural environment that it sustains are valuable and increasingly scarce and fragile. Since water is essential for mans survival and socio-economic betterment as well as for maintaining sustainability in the process of development, it is important that it is harnessed and managed in an integrated and equitable manner so as to meet the demands of competing water users and water use sectors. Demand on water is increasing and conicts are emerging at regional as well as at national level . In India, there is a mismatch between the endowment of natural resources and the population to be supported. Of the 4000 billion cubic meters (BCM) of annual rainfall, average runoff generated is 1869 BCM. Due to various constraints about 1122 BCM of water can be put to benecial use of which 690 BCM is from only surface water 432 BCM through groundwater. Out of 690 BCM of surface water, so far about 213 BCM of storage have been built through major irrigation projects. Another 184 BCM of storage are under construction. Similarly out of 432 BCM of ground water resources, about 360 BCM of groundwater is expected to be available for irrigation out of which present usage is limited to 135 BCM (GOI,2010).

Water Resources Management Issues

Planning and development of water resources related aspects need to be governed by national perspectives. India is endorsed with a large network of 12 major river basins covering 256 mha, 46 medium river basins covering about 25 mha besides other water bodies like tanks and ponds covering 7 mha with the ultimate irrigation potential of 140 mha. The surface water potential has seen developed to an extent of 37%, and that of groundwater to 38%. With the everincreasing population, there is staggering increase in water requirements for agriculture, domestic and industrial sectors. Projections show, in the coming decade there may be acute water crisis in terms of availability and supply . The problems associated with the water resources development are varied and diversied. They are viz. spatial and temporal variations in water availability of the country, falling per capita water availability of the country, rising multi-sectoral water demand for food production, domestic needs, energy generation etc., complex nature of command area problems, low efciency of irrigation systems, water logging and salinity problems of irrigated areas, signicantly large gaps exist in the ultimate irrigation potential, creation and the utilization, rapid rate of reservoir sedimentation,

in certain regions, over-exploitation of the ground water resources, deteriorating quality of the water resources due to urbanization, industrialization etc. (Dhawan,1993). To address the above issues, it calls for great challenges in the best use of available water resources through surface water capture and storage, long distance conveyance and inter-basin transfer, ground water exploitation, watershed management, conjunctive use of surface and ground water (GOI, 1999). Water Resources Management - IRS Satellite System Perspective Managing water resources is a major challenge for the country. Water resources development calls for addressing the key issues of storage, conservation and subsequently utilization. Towards evolving comprehensive management plan in suitable conservation and utilization of water resources space technology plays a crucial role in managing countrys available water resources. Systematic approaches involving judicious combination of conventional ground measurements and remote sensing techniques pave way for achieving optimum planning and operational of water resources projects. The synoptic and repetitive coverage provided by IRS satellites has effectively complemented the conventional data for monitoring the progress and impact of the above projects. The steady ow of data from Indian Remote Sensing Satellites IRS-1A, 1B, P2, 1C, P3, 1D and P4 (Oceansat) have facilitated operationalising many application areas under the aegis of National Natural Resources Management System (NNRMS). Space technology application capabilities are being further enhanced by the series of new satellites with increasing spatial, spectral and temporal resolution. The latest of IRS series of satellites viz. the Resourcesat (IRS-P6) offers multi spectral data at a resolution of better than 6 m from LISS-IV and large area coverage with high repetitivity from AWiFS payloads and the Cartosat-1 has a PAN camera of 2.5 m resolution with stereo view and Cartosat-2 with 1.0 m resolution panchromatic images. Thus, the IRS constellation has become one of the most versatile remote sensing satellite series, offering wide range of data and services to meet the variety of applications. Thus multiple satellites concurrently in operation and steer able sensor systems enable dynamic coverage of specic areas. In complement to these developments in the space segment, signicant improvements have taken place in the ground segment, paying the way for planning, execution and monitoring of water resources projects of different magnitudes. Institutional Arrangements for Use of Remote Sensing for Water Resources Management In the country, the space technology application activities are co-coordinated the National Natural Resources Management System (NNRMS) established by the Planning Commission, for which Department of Space (DOS) is the nodal agency. NNRMS has been facilitating inventory of the countrys natural resources for its optimal utilization through application of Remote Sensing technology in conjunction with the conventional methods. The Planning Committee of NNRMS (PC-NNRMS) is the apex of NNRMS, which is chaired by Member (Science) Planning Commission with Secretaries of Ministries/Departments as members. Ten thematic Standing Committees support the PC-NNRMS, of which Standing Committee on Water Resources (SC-W) is one. Standing Committee on Water Resources (SC-W) since its formation in the year 1984 has been providing the essential thrust and guidelines for many important projects in water resources sector

87

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

frequent oods severely affecting the development process in ood prone areas, recurring drought

and operationalisation of remote sensing using IRS satellite data. Ministry is periodically implementing the plan scheme on remote sensing applications in water resources development and management, through which several important remote sensing projects are taken up in the elds of reservoir sedimentation, irrigation command area monitoring, river basin development , creation of national digital water data bases , development of national water resources information system, snow & glacier studies. Use of high resolution remote sensing data for monitoring the irrigation infrastructure and feasibility studies for interlinking of river basins are major recent initiatives. Development of a national Water Resources Information system (India WRIS) is underway jointly by the Department of Space and Ministry of Water Resources (MoWR). Further, we have today an operational space-based disaster watch system in the country that monitors all ood events, cyclones and provide timely advisories to all stakeholders. Remote sensing data is also being extensively used in strategic applications. Ministry of Water Resources, Govt. of India is also planning to institutionalise the space technology applications through establishment of National Water Resources informatics centre (NWRIC) with databases and networked data banks, integrating and strengthening the existing Central and State level agencies. This initiative is intended to develop and adopt improved methodological framework for measuring/monitoring the progress of the countrys water sector towards a sustainable development and management.

Addressing the Water Resources Management issues through Space Technology

The Indian Space Programme has been characterized by a vision to use space technology for sustainable development of natural resources including water resources. Indias Earth Observation (EO) satellites, the Indian Remote Sensing Satellites (IRS) 1A to P6, are tailored to meet the critical information needs of water and allied sectors. Since the modest beginning of surface water inventory, the remote sensing application scenario has witnessed a phase transition from resource mapping to decision-making. Remote sensing inputs are being used for number of applications, such as rainfall estimation, snow and glacier studies leading to snow melt runoff forecasting, irrigation water management and identication of potential irrigable lands, identication of salt affected and waterlogged areas in the command areas, reservoir sedimentation, watershed management, disaster management, water quality assessment, ground water assessment and prospecting, planning and implementation of developmental activities, irrigation infrastructure, disaster management and hydrological modelling, etc. Remote sensing has thus become an important tool in evaluation of the physical attributes of water and land resources in the country. Irrigation Water Management and Command Area Development In India the ultimate irrigation potential of the country has been estimated at 139.91 mha. Full development of irrigation potential is expected to be achieved in next two decades. Optimum development of river systems, including inter basin transfers could increase the ultimate irrigation potential by another 35 mha.. While enormous irrigation potential has been created at huge cost, the gap between created potential and utilisation is signicantly large (around 14 mha ). One of the major draw backs of the water utilisation pattern in India has been the poor water use efciency. Adverse impacts of salinity and water logging as a result of over irrigation and inadequate drainage are also limiting factors (GOI,2002). To monitor the impact of Command Area Development (CAD) activities on major irrigation projects since 1985, Ministry of Water Resources has requested DOS to conduct satellite based performance evaluation study of 14 selected irrigation projects falling in 5 states. Since then, the technology is being applied in many command areas for monitoring and evaluation purposes ( NRSA.1998). The temporal and spatial analysis of satellite data has indicated problem pockets of poor performance. Diagnostic analysis supported by farmer surveys has identied causative factors for corrective management. Satellite data has helped in these projects in identifying cropping pattern, the extent of unauthorised

plans. Spatial analysis of crop sowing periods and crop condition assessment have thrown up policy issues of relevance to irrigation scheduling, canal maintenance and agricultural productivity. Satellite data has also been analysed to map current status of water logging and soil salinity in all the irrigation projects. Analysis of satellite data has also helped to evaluate effectiveness of reclamation programmes by monitoring extent and severity of soil limitations through the years. Multi temporal Resourcesat 1 &2 AWiFS data was used to capture irrigation dynamics in saving around 15-20% of water. Thus, AWiFS data from IRS satellite has potential for development of real-time decision support systems for irrigation water management. Irrigation Infrastructure Mapping The advent of high-resolution remote sensing satellites providing information over relatively small areas with greater details has opened new era of remote sensing applications. The meter to submeter pixel resolution satellite data dramatically improved the feature identication. High resolution satellite imagery can provide a rapid, high quality data source for capturing existing irrigation network and associated infrastructure details. Accelerated Irrigation Benefits Programme (AIBP) Accelerated Irrigation Benet Program (AIBP) was launched by Government of India during 1996-97 to provide nancial assistance to State Governments with the aim of speeding up the implementation of on-going irrigation / multi-purpose water resources projects. Monitoring of the projects covered under the AIBP is periodically done by the Central Water Commission / Ministry of Water Resources with the help of its regional ofces situated all over the country. In the absence of real time information during construction, the conventional monitoring is done through discussions with eld authorities and random eld checks. The monitoring reports with non-spatial information generally lack in the synoptic view of the critical gaps and the quantitative progress achieved in irrigation potential creation. NRSC, ISRO has successfully

demonstrated the assessment of irrigation potential created using inventoried irrigation infrastructure from high resolution satellite data (Figure 1). Initially, 103 projects (Phase I-53No, Phase II -50No.) were carried out by NRSC/ISRO (NRSC,2010). Based on the utility of such studies, Central
Fig. 1: Irrigation infrastructure and irrigation potential assessed using Cartosat 1 in Gundlakamma Project

Water Commission (CWC) decided to internalise the use the satellite data and web services through ISRO-Bhuvan for monitoring

150 ongoing projects at six months interval (pre monsoon & post-monsoon from the year 2013). This would result the generation of reports with information on physical progress of the canal network vis--vis design, critical gaps in canal stretches and status of hydraulic connectivity before they visit the eld. 89

N N R M S

B U L L E T I N

real time in Hirakud command area resulting in developing optimal irrigation scheduling strategies

M A R C H

2 0 1 3

irrigation and poor recovery of water rates, thus providing inputs for changes in policies and operational

ISRO Bhuvan web portal would facilitate the hosting of satellite images by the end of September-December and Feb-May time windows for all the projects as desired by CWC. Customized tools for online mapping, report generation and map generation are developed for facilitating near real time monitoring the progress of AIBP funded irrigation projects. CWC Ofcials will be using this tool before going for eld visit and summarize the status useful for annual/sub annual monitoring of the projects and appraise the status to the Govt. of India. The report can be generated online with help of eld data entered by the project authorities / central water commission regional ofces along with the satellite data derived information on irrigation infrastructure. Monitoring of Surface Water Bodies According to the land use statistics, the area under tank irrigation reached an all-time high of 4.78 mha. in 1962-63 and progressively came down to 3.07 mha in 1985-86. It is clear that maintenance of tanks is being neglected due to which about 17 lakh ha. area has gone out of irrigation in 15 years. Most of them need substantial investment for renovation and improvement. It is necessary to restore the storage capacity of water bodies with the purpose of recovering their lost irrigation potential( Himanshu Thakkar,1999). As a part of evaluation of 742 tanks spread in 6 states in 9 districts under NPRRR ( National Project for Repair, Renovation and Restoration),multi-spectral satellite data from (IRS P6/1D/1C) LISS III sensor (24 m resolution) acquired during 2004-05 and 2007-08 years were used to derive season-wise performance indicators. High-resolution multi-spectral data from Resourcesat-1 LISS IV sensor (5.8 m resolution) were used for precise identication of each tank command jurisdiction. The annual irrigation utilization of 742 tanks increased from 76414 ha during 2004-05 to 98138 ha during 2007-08 with an increase of 28.4%. Out of the total 742 tanks/water bodies, 82 tanks (73%), show improvement from 2004-05 season to 2007-08 season, 96 tanks (13%) show decrease in performance and no signicant change observed in 102 (14%) tank commands. NRSC has developed a knowledge based hierarchical algorithm for quick processing and analysis of satellite data of Resourcesat AWiFS and LISS III data for providing information on surface water area, which in turn can be used to for rapid inventory and monitoring of water bodies which are useful for hydrological modeling and climate studies, water surface dynamics at regional/ river basin level regional level, ood inundation mapping. Reservoir Sedimentation Many reservoirs built at huge investment are undergoing rapid silting and loss of storage capacity and consequent reduction in economic life of reservoir. The analysis of sedimentation data of Indian reservoirs show that the annual siltation rate has been generally 1.5 to 3 times more than the designed rate and the reservoirs are generally losing capacity at the rate of 0.30 to 0.92 per cent annually. Conventional hydrographic surveys to reassess reservoir capacity are time consuming and costly. Satellite remote sensing techniques have been demonstrated to be cost and time effective in many studies conducted by CWC, While this technique helps in revising capacity table between minimum and maximum draw down level observed in satellite data. It is suggested that while satellite data can be analysed to estimate the storage reduction in the live storage region, conventional hydrographic surveys can be employed for the balance of capacity reduction in the dead storage. A national program on satellite remote sensing based sedimentation survey of 124 major reservoirs across the country monitored by Ministry of Water Resources, Govt. of India. All the above reservoirs are monitored once in ve years. many of the state engineering research laboratories and hydrography divisions are now routinely use satellite data for sedimentation surveys every year. The deployment of a knowledge based hierarchical algorithm developed by NRSC is being planned to integrate into image processing chain for rapid assessment of periodic (weekly or fortnightly) reservoir storages across country.

Hydro-power potential of the country is 1,48,000 MW. Out of this, nearly 20% hydropotential has been developed and 10% is under various stages of development. In order to exploit the balance hydro-potential located in 450 sites , it is necessary to intensify technical planning (survey, investigation, design), and programme implementation. The potential of space technology was amply demonstrated during the preliminary ranking studies stage of the proposed hydroelectric sites in Indus basin by NRSA in the year 2001 for Central reports, NRSC has conducted initial environmental studies through satellite remote sensing technology in respect of 45 proposed hydro-power sites in Arunachal Pradesh, Himachal Pradesh, Jammu & Kashmir and Madhya Pradesh States for NHPC, HPSEB during 2003 -04 . The high resolution satellite data ( IRS 1C/1D PAN+LISS-III merged data) was used to estimate the Land use / Land cover area in the proposed dam submergence area and its immediate environs of hydro-power sites mostly located in inaccessible areas. Also, identication of the infrastructure details like roads, bridges, settlements etc. was carried out. Initial environmental assessment using submergence area statistics covering rehabilitation & relocation aspects and impact of proposed hydro-electric projects on national parks and sanctuaries using proximity analysis was carried out. many such similar feasibility assessment studies were conducted by ISRO and other remote sensing centres for Government and Public sector companies. Space technology inputs were also used in environmental assessment of proposed Tamanthi hydro-electric project on Erawadi River in Myanmar for NHPC, which is a bilateral project between Ministry of External Affairs, Govt. of India and Govt. of Myanmar. Snow and Glacier Studies Snow acts as a water reservoir. Hence, assessment of its possible effect on snowmelt runoff is essential for planning and management of various multipurpose projects on snow fed rivers. Retreat of glaciers is the cause of concern for the perennially fed Himalayan river system. Hence there is a need to develop long-term database on snow glaciers and melting patterns of snow in the Himalayan river basins to check climate signals and monitor temporal and spatial variations of discharge. Remote sensing images provide ample information on the status of the glaciers and dynamic changes over time. Inventory of glaciers in the Himalayas was carried out using satellite data on 1:250,000 scale. Glaciers larger than 0.56 sq.km were mapped. Space technology is used for snow cover mapping both permanent and seasonal, snowmelt pattern assessment, snow pack characterization, accumulation and ablation areas, which are being used developing snow melt run-off forecast models. Ministry of Environment and Forest in association with Department of Space initiated a project to study the Snow and Glaciers in the country including glacial inventory of entire Himalayas on 1:50,000 scale, snow cover monitoring on 1:250,000 scale every 10 days and glacier retreat & mass balance studies for selected glaciers, leading towards a development of Himalayan snow and Glacial information System. The inventory of glacial lakes and water bodies in the Himalayan region of Indian river basins carried out by NRSC using Resourcesat1 AWiFS images in 2009 shows presence of 2028 glacial lakes and water bodies within the Himalayan region of Indian river basins. Out of these, 503 are glacial lakes and 1525 are water bodies. Brahmaputra basin part of the Himalayan region contains 294 glacial lakes and 1099 water bodies whereas Indus basin has 31 glacial lakes and 321 water bodies. Monitoring of Electricity Authority (CEA), Ministry of Power, Govt. of India. As part of preparation of pre-feasibility

91

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Hydro Power Development

the glacial lakes and water bodies (of size more than 50 ha in water spread area) was carried out for the months from June to October during the years 2011 and 2012 using Resourcesat-1/2 AWiFS images. The seasonal (April June) snowmelt runoff at Bhakra Reservoir, derived from satellite monitoring of snow covered area, has been forecasted with less than 10 per cent error for use in Bhakra Beas Management Board (BBMB) during the last decade. Simlarly, NRSC has developed seaonal snow meltrunoff models for ve himalayan rivers namely Chenab (upto Premnagar), Beas (upto Bhuntar), Yamuna (upto Hatnikund), Bhagirathi (upto Uttarkashi), Alaknanda (upto Rudraprayag) and Sutlej (upto Bhakra) in the year 2012 .Satellite derived snow cover, surface temperature and eld observed hydromet data using energy balance approach to forecast runoff at the basin outlets comprising snowmelt, glacier melt, runoff due to rainfall and base ow components. Presently work is under to progress to develop (10 daily or 16 daily) short term runoff forecasting for the above ve river basins. Interlinking of Rivers The feasibility of river interlinking involves the various aspects ranging from topography, geology, hydrology and environment. The information on the above segments is essential towards establishing the links as well as irrigation planning in the command areas. In this regard, space systems have an important role to play while addressing them. Space inputs to feasibility, detailed project, implementation strategy, monitoring and evaluation studies help in harmonizing them and lead to an optimal solution. This was amply demonstrated by NWDA in its preparation of one of the pre-feasibility study reports of Krishna Pennar link between Nagarjunasagar reservoir in Krishna basin and Somasila reservoir in Pennar basin, which is one of the components of peninsular, inter basin water transfer. The 3-dimensional terrain modeling with the state-of art sensor data products from the Airborne Laser Terrain Mapper (ALTM)/ and high-resolution satellite data (Cartosat 1 stereo data, is essential for precise identication of the alignment of canals and other structures. The ALTM technology was successfully demonstrated in Godavari-Krishna link canal project. NRSC has conducted a study for NWDA to assess feasibility of proposed projects in Upper Betwa basin as part of Ken-betwa interlink project .Cartosat-1 stereo pair data of 2.5 m spatial resolution (40 scenes) was used for generating the Digital Elevation Model (DEM) . Height control survey in MSL was affected and DEM was used to generate 10 m contours and interpolated to derive 1m contours over a 30 m grid image. Digital integration of land use/land cover map( prepared using IRS P6 LISS IV data of 5.8 m spatial resolution) and 1m contours facilitated generation of statistics for each project site. Submergence area statistics, in terms of major land use / land cover categories, at varying proposed FRLs (7-10 for each project site) were generated at 1m interval. Area that can be commanded for the proposed MDDL (Minimum Draw Down Level) was generated to estimate the gross command area that can be provided with irrigation service under the proposed project. The capacity under various FRLs (7-10 each) was also quantied and preliminary Area - Elevation capacity curves for all the proposed reservoirs were prepared. The study assisted NWDA to evaluate various proposed alternatives and nalize design parameters on a scientic basis and will help in optimal utilization of water resources in Upper Betwa basin. Groundwater Prospects Zone Mapping There is a need for scientic planning of ground water development in different hydrogeological set ups to evolve effective management practices. Remote sensing data helps in delineating the potential groundwater occurrence zone in less time and cost effective manner than the conventional methods. The thematic layers on geology, geomorphology, drainage, lithologic, landform, structural, slope, soil and land use generated from satellite data, are essential in identifying the groundwater prospect zones..

of Inida in 1999), preparation of ground water prospects map has been carried out on 1:50K scale using high resolution satellite data (IRS LISS-III data) for the entire country. The main objective of this project has been to enable scientic source nding of drinking water in all the Non-Covered (NC) and Partially Covered (PC) habitations (around 2 lakh in number) and to identify suitable zones for putting recharge structures. Hydrological Modelling & facilitate forecasting hydrological events, such as ood disasters. Various model approaches have been adopted for this purpose ranging from simpler lumped and conceptual catchments models to complex distributed and physically based models. Earth Observation (EO) data, when used jointly with insitu data, can provide an essential contribution for the creation of spatially distributed parameters relevant for hydrological studies and models. Repeatability of observations allows the generation of a time-series of dynamic terrain parameters and provide capability to quantify and forecast the hydrological variables and water balance components. The improvement of the quantity, quality and the resolution of remote sensing data over the last decades enhanced the integration remote sensing data into hydrological studies. Assessment of Basin-Wise Water Resources Different commissions, agencies, researchers have estimated water resources of the country using different approaches. Reassessment of Average Annual Water Resources Potential (1993) is the most recent and authentic study done by Central Water Commission. CWCs recent study was done for the period of 1967 to 1985 for many river basins. All these studies were based on the observed ows at terminal sites and upstream abstractions for irrigation and domestic consumptions. Limitations of these studies are; limited eld data on abstractions, lumped approach in estimation at terminal sites of the basins, no meteorological parameters (such as rainfall evapotranspiration) are taken into consideration, and no mechanism for cross validation (NRSC, 2009). Basin scale water wealth quantication by transformation from presently adapted basin terminal gauge site runoff aggregation to distributed hydrological modelling approach has been demonstrated in a joint pilot study by ISRO and CWC in Godavari and Baitarani -Brahmani river basins. The study was aimed at computing mean annual water resources in during the last 20 years (1988-89 to 2007-08) and to compute availability of water resources during extreme wet & dry rainfall conditions in the basins. These landuse grids were integrated with soil textural, digital elevation and command area grids to compute hydrological response unit grids. In this study, a newer approach was adopted in computing monthly potential evapotranspiration by incorporating landuse coefcients derived from the satellite data. Groundwater, reservoir ux and domestic water consumption were computed using the spatial groundwater, reservoirs level and demographic data respectively and integrated in the model environment in computing runoff. The model was calibrated and validated using observed discharge data at various prominent gauge stations of the respective basins by keeping land use coefcient grids as variable. It is found that computed runoff is very well matching with the eld data with more than 90% accuracy (NRSC,2011). This methodolgy was ratied by national committee for re-assessment constituted by Ministry of Water Resources, Govt. of India and the study is being extended to all river basins in the country during 12FYP.

93

N N R M S

B U L L E T I N

Hydrological models are valuable tools that provide quantication of water balance components

M A R C H

2 0 1 3

Under Rajiv Gandhi National Drinking Water Mission (RGNDWM), (launched by Government

Flood Forecasting & Inundation Modelling - Development of Flood Early Warning Systems
NRSC has successfully developed flood forecast model for Godavari basin using an open source hydrological modeling softwares like HEC-GeoHMS and HECHMS. Various space based inputs and geospatial database such as landuse/landcover, digital elevation model, soil textural data, and non-spatial data base such as hydrometeorological data and gauge discharge curves were used in the study. The model was calibrated with the hydro-meteorological data of 2000 and validated with the data of 2000
Fig. 2: Hydrological description of the study area

(years 2000 and 2001 are heavy ood years for the Godavari). Subsequently, model was, implemented during the 2010,2011 and 2012 ood season using the real-time 3 hr interval hydro-meteorological data by CWC and NRSC to provide ood forecast hydrographs are computed the ood forecast stations namely, Perur and Koida that covers the Badrachalam and coastal floodplains (Figures 2 & 3). Similarly, NESAC has developed FLood Early Warning System (FLEWS) for a few severely ood affected tributaries of Brahmaputra. It integrates satellite based rainfall estimates over the inaccessible Himalayan watersheds with the run-off models in a GIS environment and gives spatial and temporal dimensions of an impending ood event.

Fig. 3: Computed and observed ood hydrographs at Koida station

Flood inundation simulation studies for part of Sabari tributary of Godavari river from Konta to Kunavaram (35 km) stretch in Godavari River basin were carried out using discharge data, land use/land cover and ALTM DTM and the results were validated with satellite data. The river geometry data extracted using HEC-GeoRAS, together with the discharge data and other parameters, spatial ood modeling has been carried out for peak ood situation using HEC-RAS, 1-D hydraulic model. Hourly water level data was used to generate different ood inundation patterns. Flood inundation is simulated for the year 2010 and the maximum ood inundation extent simulated by the model was compared with the corresponding actual inundated area obtained from Radarsat image. The results obtained from the model for peak ood situation were in well accordance with ood inundation observed from corresponding satellite data. Ministry of Water Resources giving impetus to the ood inundation and forecasting modelling, which is very important non-structural ood damage control measure during the 12 FYP. Towards this the above pilot studies would act as capacity building for various central and state organisations including CWC. CWC has initiated measures to augment infrastructure and capacity building with the help of NRSC as part of internalising the technology.

The National Water Policy (GOI, 1987, 2002 & 2012) stipulates that the prime requisite for resource planning shall be a well developed information system consisting of scientifically designed data bases for data processing capabilities. Department of Space and Ministry
Fig. 4: The screen shot of India-WRIS web portal home page

of Water Resources jointly under NNRMS umbrella have developed a national Water

Resources Information system (India WRIS: www.india-wris.nrsc.gov.in) to provide a single window solution of all water resources and related data in a standardized GIS format in national framework. India WRIS web portal formally launched on April 2011.. India-WRIS will provide a comprehensive, credible, and contextual view of Indias water resources data along with allied natural resources data and information. It will allow users to search, access, visualize, understand, analyze, look into context and study spatial patterns . The inventory is having a Geographic Information System (GIS) which provides exible information on water resources that can be used based on political/administrative and on the hydrological boundaries (river basins level to watershed level). Each inventoried water supply facility in each watershed has GIS information and this will allow in the near future to estimate the annual water consumption/abstraction or withdrawal for different sectoral water use at a river basin level, a data that remains for a long time as a gap in determining the water balance at river basin level. The web portal is being constantly upgraded and new and new data sets are being integrated periodically. India-WRIS Web GIS Version 3..0 has 12 major information systems, 35 sub-information systems, 87 spatial layers and temporal data of 5-100 years. The portal has 6 modules for visualization, information catalogue, users services, input data built-up, module 2 share success stories and created WRIS. The complete content as envisaged would be available be the end of December 2013.

Future Thrust Areas

With the improved eld observational networks and Remote sensing and GIS capabilities, the focus would be on scientic assessment, development and monitoring the available water resources. As CWC has made operational the India-WRIS web portal, priority is given for capacity building for smooth transfer of technology for institutionalization and setting up of National Water Resources Informatics centre (NWRIC) to develop and adopt improved methodological framework for measuring/monitoring the countrys water resources for sustainable development and management. As the country has embarked upon creating national digital water data bases and with availability of high resolution satellite data, the stage is now set for developing locale-specic strategies for beneting at the grass-root level. Recent studies have brought out the adverse impact of climate change on water resources in the country. Realistic estimation of water resources in spatial and temporal domain is required. Hence, in the coming years efforts would be made to address the above issues with advanced technologies. Periodic Basin-wise water resources assessment, ood forecasting and ood inundation modeling of all ood prone areas, periodic 95

N N R M S

B U L L E T I N

improving both the quality of data and the

M A R C H

National Water Resources Information System (IndiaWRIS)

2 0 1 3

monitoring of Irrigation projects for monitoring Irrigation infrastructure and irrigated area assessment, development of DSS for real-time irrigation water management, micro level ground water aquifer mapping and modeling for sustainable management of groundwater resources, Snow cover dynamics monitoring and snow melt runoff forecasting, glacial inventory & monitoring and climate change impact are some of the key areas requiring critical inputs from satellite remote sensing in the near future

Conclusion
Since, the modest beginning of surface water inventory with the launch of IRS-1A satellite in the year 1988, the remote sensing application scenario has witnessed a phase transition from resource mapping to decision-making. Remote sensing has thus become one of the most important tools for evaluation of the physical attributes of water and land resources in the country. A number of case studies on command area development, canal alignment, irrigation performance evaluation, etc., have proved beyond doubt that integration of remote sensing and conventional approach signicantly decrease the cost and time involved as well as, improve the steadfastness. The various issue related to topographical surveys, irrigation infrastructure, water resource assessment in both rainfed and snowfed baisns, information on command area expansion, planning of new storage reservoirs, stabilizing existing enroute command areas, reservoir sedimentation, glaciers, river conguration, ood forecasting and ood inundation studies, etc. have been suitably addressed through satellite data from IRS series of for the last two and half decades. The water and water related data bases generated using IRS satellite data over the last 25 years under various programmes under NNRMS umbrella were used together with large hydromet and environmental data from CWC, for developing a standardized national information system viz., India-WRIS jointly by ISRO and CWC. India-WRIS is a comprehensive water data and information gathering and monitoring system that delivers water related information which can be used effectively in decision making in the country. Space borne multi spectral measurements have in some cases replaced ground based observations and in others complemented at varying levels. Improved spatial, spectral and temporal resolution data from present IRS satellites together with future remote sensing satellites provides unique opportunity towards comprehensive monitoring of water resources dynamics in the country.

Acknowledgements
Authors sincerely place on record the contributions of all the concerned scientists across various ISRO centers who have executed several projects, which have been mentioned in the manuscript and their contribution is acknowledged. Authors also place on record various ministries viz. Ministry of Water resources, Ministry of Environment and Forest, Ministry of Drinking Water and Sanitation, Ministry of Rural Development for the valuable support in executing the several projects mentioned in the manuscript.

References
Dhawan, B. D. (1993). Indian Water Resource Development for Irrigation: Issues Critiques and Reviews, New Delhi: Commonwealth Publishers. GOI, (1987). National Water Policy, Ministry of Water Resources, Govt. of India, New Delhi. GOI, (1999). Water Resources Development Plan of India: Policy and Issues, New Delhi: National Commission for Integrated Water Resources Plan, Ministry of Water Resources, GOI. GOI, (2002). Report of the Steering Committee on Irrigation for the Tenth Five Year Plan (2002-2007), Planning Commission,GOI. GOI, (2002). National Water Policy, Ministry of Water Resources, Govt. of India, New Delhi GOI, (2010). Annual Reprt,2009-10, Ministry of Water Resources ,GOI. www.wrmin.nic.in GOI, (2012). National Water Policy, Ministry of Water Resources, Govt. of India, New Delhi

Commission on Dams. NRSA.(1998). Project report : Performance Evaluation of 13 irrigation command areas in 5 States, Study commissioned by, Ministry of Water Resources, GOI. NRSC. (2010). Project Report on Assessment of irrigation potential through mapping of irrigation infrastructure using high resolution satellite data in 53 AIBP Projects. NRSC, (2009). Water Resources Assessment the National Perspective: A Technical Guide for Research and Practice (August 2009). : NRSC-RS&GISAA-WR&OG-WRD-Oct 09-TR98. NRSC, (2011). Assessment of Water Resources at Basin Scale using Space Inputs (December 2011): NRSC / RS&GISAA / WRG / WRD / NRSC-CWC Pilot Study / R1 /Dec 2011/ TR 369. NRSC. (2012). Project Report on Groundwater Prospects mapping Under Rajiv Gandhi Drinking Water Mission. Study commissioned by Ministry of Rural Development, GOI.

97

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Himanshu Thakkar, (1999). Assessment of Irrigation in India. contributing papers to the World

ADvANCES IN SAtELLItE OCEANOGRAPHY AND MEtEOROLOGY

Raj Kumar, Rajawat AS, Rashmi Sharma and Kishtawal CM Space Applications Centre, Ahmedabad - 380 015, India Email: rksharma@sac.isro.gov.in

Introduction
The physical environment of Indian sub-continent and surrounding oceans has unique features. Along with the seasonal weather changes, associated changes in the surface and subsurface oceanic thermal and dynamic behavior determine the overall physical environment of the whole region such as summer monsoon and periods of intense activities such as tropical cyclones. Almost all human activities, in this region, are critically dependent on the variability of this physical environment since a major critical component of the economy is agriculture and marine activity (like shery & transportation). Understanding the variability and prediction of these phenomena at different time scales, therefore, become most signicant to farmers, shermen, livestock producers, and professions working in health, economic, transport, offshore exploration and navigation, and those in the insurance industry. Keeping this in mind, Meteorology and Oceanography have been of keen interest to ISRO since inception of remote sensing programme in early 70s. The major goal of the research is towards application of space technology for facilitating improvement in weather, ocean and climate forecast at various spatio-temporal scales through retrieval and data assimilation in model. With the launch of multifrequency Satellite Microwave Radiometer (SAMIR) onboard experimental satellites Bhaskara-I and II in 1979 and 1981 respectively, data from these satellites provided the beginning to oceanic and atmospheric sciences activities. Since then, the use of spaceborne data for the retrieval of geo-physical information and studies of underlying processes in oceanography and meteorology took a major stride forward with the availability of data from international satellites such as SEASAT, ERS, TOPEX, JASON, QuikSCAT, ENVISAT, NOAA etc. The impetus, however, was provided with the launch of IRSP3, launched in March 1996, carrying Modular Opto-electronic Scanner (MOS) onboard and Oceansat-I, launched in 1999, carrying a Multi-frequency Scanning Microwave radiometer (MSMR) and an Ocean Colour Monitor (OCM) demonstrating end-to-end experience in developing the capabilities at national scale. Major emphasis was placed on operationalisation of retrieval of geo-physical and biological parameters from these sensors, ingest in the dynamic models of atmosphere and ocean and to demonstrate the applications in terms of real time usage for weather and ocean state forecast. This also provided an opportunity to integrate the efforts being made, on national scale, by operational agencies and academic institutions.

Ocean Observations
The feasibility of measuring physical oceanographic parameters such as sea level, ocean wave heights and wind speeds from space was demonstrated more than thirty years ago with the launch of GEOS-3 and the SEASAT in 1970s. Since then, algorithms have been rened, and

space agencies have to be carefully quality controlled and calibrated. One of the major applications of satellite altimeter is to provide sea surface height. The efforts are being made to retrieve geophysical parameters for open ocean and Indian coastal regions with the upcoming SARAL-AltiKA data. For derivation of Sea Surface Height (SSH) within a few cms accuracy, accurate measurements of time difference as well as different atmospheric and other sea state corrections are needed. The EM bias correction is believed to be the source of one of the largest uncertainties in altimetric SSH observations. Efforts were made to characterize the impact of such corrections on the altimeter data. Using wave and empirical corrections based on wave slope and wave age information over a signicant range of parameters. The other important parameters related to the surface waves are wave period, swell height and extreme waves. Swells are the long period waves generated due to winds, which propagate to the other oceanic regions without the inuence of local winds. Extreme wave analysis is important for offshore industries and ship routing. Since these parameters are not directly available from the satellite data, efforts have been made to derive the swell height and wave period from the satellite altimeter data using different techniques. The climatology of all these parameters have also been generated using 10 years of satellite altimeter data, which is useful for offshore engineering as well as ship routing studies. Using the altimeter data of 10 years and numerical wave model analysis, extreme wave analysis has been carried out to predict 50 and 100 years wave heights. Both the satellite data and model analysed results show similar trends. As the winds are major forcing for all ocean models, the accurate observations and prediction of surface winds is quite important. The frequency of spaceborne observations of surface winds had increased tremendously after the availability of QuikSCAT as well as ASCAT scatterometer data. After 10 years of regular availability of wind vectors by QuikSCAT, there was a sudden gap due to problem in QuikSCAT. However with the ISROs launch of Oceansat-2 Scatterometer on September 23, 2009 with repetivity of 2 days and swath of ~1800 km, the surface wind data availability has been continued. The Oceansat-2 scatterometer is a Ku-band pencil beam radar, capable of measuring backscattering coefcients at vertical and horizontal polarizations in swaths of 1800 and 1400 km. Initially algorithms were developed for the retrieval of ocean surface wind vector from SeaSat, ERS-1/2
Fig. 1: Scatterometer derived operational and high resolution winds

and ADEOS scatteroemters. The near real time, ocean surface winds at 50 x 50 km or better spatial resolution as well as winds at

ner spatial intervals have been retrieved for Oceansat-2 scatterometer and are available for major atmospheric and ocean applications (Figure 1). Extensive calibration and validation exercises for the Oceansat-2 scatterometer have been carried out. The experiments clearly demonstrate that the ocean surface wind data being made available by the Oceansat-2 Scatterometer is well within the mission specications. Assimilation of surface wind vectors is known to have signicant impact on accuracy of the weather forecast models. In-house assimilation experiments have shown the positive impact of 99

N N R M S

B U L L E T I N

buoy and wave model data, a good agreement has been found between existing theoretical correction

M A R C H

2 0 1 3

extensively validated against in-situ data. However, the routinely processed data provided by the

Oceansat-2 scatterometer on the weather predictions. These wind predictions feed into ocean state forecast models for continuous monitoring and prediction of ocean surface waves and circulations. A directional wave spectrum is a compact means of presenting the important physical characteristics of a given wave eld. Under the RISAT-1 Utilisation programme, various SAR datasets have been analysed to derive 2-Dimensional ocean wave spectra. The retrieved ocean wave spectrum, on comparison with buoy data and the coastal wave model results showed that SAR is able to estimate swell waves very well. High resolution ocean surface winds have also been retrieved using many ENIVSAT datasets and comparisons have been made with numerical models, QuikSCAT & ASCAT data. For coastal depth estimation, the algorithm is developed using EnviSAT ASAR data. Algorithm for ship detection has been developed using EnviSAT ASAR data by adaptive threshold technique and veried with knowing targets locations. These techniques are being ne tuned with larger datasets of RISAT-1, built and launched by ISRO in April, 2012.

Ocean Predictions
O c e a n S t a t e F o re c a s t ( O S F ) was identified by Department of Ocean Development as one of the primary goal. An experiment with OSF was performed to
Fig. 2: Sea level and wave height predictions using numerical models

assess the real time requirements for such work. Thus, a pilot experiment was conducted

during May-July 2001 to provide the forecast for near real time basis and get the feedback from the users. A wave forecast system was developed at Space Applications Centre. Using this system, operational wave forecast was carried out during the month of May-July 2001. Aim of the study was to assess the forecast capability of the numerical models for Indian Ocean region. One of the best ways of using satellite observations operationally appears to be through forcing of the geophysical parameters and assimilation of the parameters in numerical models. Numerous experiments to study the impact of satellite data have been performed using Ocean General Circulation Model for global as well as Indian Ocean region. Impact of satellite derived parameters such as SST, precipitation, chlorophyll winds and salinity data on ocean variables has been studied through their use in ocean models. Using the model analysis, ocean processes ranging from interannual (eg., Indian Ocean Dipole) to intraseasonal scales in the oceans have been studied. For wave model, various experiments were carried out to show the sensitivity of the model to forcing winds and impact of swell in Indian Ocean. For the prediction of waves in the Indian coastal region, coastal model was nested with global wave model and experiments were performed to demonstrate the wave predictions in operational model. These models are being utilized for operational predictions. Figure 2 depicts the typical sea level and ocean surface waves simulated by these models. It has been seen that there are three well-dened swell dominant zones in the tropical areas of Pacic, Atlantic, and Indian Oceans. This emphasizes the fact that, for accurate wave prediction over the Indian Ocean, accounting for swell part of the waves is crucial and state of the Indian Ocean is signicantly modied by the swell waves and should be critically accommodated for predicting the sea state, in particular, for the coastal applications. The present state of the ocean can be determined by in-situ observations as well as observations from space platforms. The best possible ocean state analyses can be generated using the in-situ measurements at a specic location. In the last few years Argo Floats have become major source of sub-surface information. However, due to vast extent of oceans, it is next to impossible to deploy in-situ instruments at each and every location. In view of the above, the spaceborne sensors are crucial for integrated ocean observation system. The major advantage of spaceborne observations is their high temporal repetivity and global coverage. Even with the combination of in-situ and spaceborne observations, the information about ocean state may not be possible at desired temporal and spatial intervals, whereas, the numerical models provide ocean state parameters at regular grids.

model schemes, mismatch remains between models predicted parameters and in-situ or satellite observations. This is mainly due to the lack of understanding of proper physics in the model and assumptions used in the model. In such a situation, the best way to reduce the errors is assimilation of satellite and in situ data in the model. There are different techniques of assimilating geophysical parameters in the numerical models. Various assimilation experiments such as identical twin experiment, assimilation using direct insertion, spreading, nudging, optimal Interpolation etc., have been performed to assess the impact of wave height, sea level, surface temperature data on the model. The projection model. Assimilation of altimeter derived wave heights and SAR derived wave spectrum have been demonstrated to show its impact in wave simulations. It has also been observed that spatial coverage of altimeter also plays an important role. The performance of assimilation scheme is much improved when spatial coverage by altimeter is increased. It is also important to note that repetivity of the altimeter also plays an important role in continuous data assimilation procedure since the impact of the assimilation may diminish in the case of less frequent observations at a single point during the assimilation cycle itself. Due to large number of coastal population, sediment transportation, upcoming industries, harbours design and operations, search and rescue operations, etc., the predictions of ocean parameters in the coastal regions has been very important. In the coastal prediction, in addition to forcing terms, the bathymetry, type of bottom and shape of coastal boundaries also play a major role. In view of that, the coastal predictions models emphasise on higher spatial resolution and importance of above effects. The prediction in the coastal regions becomes more challenging because of more complexity in processes of generation, non-linear interaction and dissipation. A number of studies have been carried out for the sea-ice extent using SSM/I, MSMR and OSCAT data. Monthly variations of the sea ice in the polar regions were studied using MSMR data, and sea ice atlases were generated for the Arctic and Antarctic regions. A study has also been carried out for the estimation of thin sea-ice thickness from AMSR-E data. The assessment of sea-ice cover changes and changes occurring in the ice-shelf surface melting has also been studied. The assessment of sea-ice variability or ice-sheet surface melting using the space-borne data has also been studied. The long term changes in the sea ice extent have been studied using QuikSCAT scatterometer data. The onset melting date has also been derived using Oceansat-2 data. based technique to utilize sea level data assimilation scheme has been developed for circulation

Meteorology
Synoptic coverage provided by satellites is ideally suited to study weather related atmospheric processes on different scales. The recent advances in satellite technology in terms of high resolution, multi-spectral bands covering visible, infrared and microwave regions have made space data an inevitable component in weather monitoring and dynamic modeling. The impact of satellite data is phenomenal in certain areas of meteorological applications such as short-range forecasts, tropical cyclone monitoring, aviation forecasts etc. With improving trend in accuracy of satellite retrievals, improvements could be carried out in models leading to improved forecasts, especially in the tropics. INSAT series of satellites have provided the most consistent operational services to the nation for more than three decades. INSAT applications program started with the launch of INSAT-1 series of satellites in early 1980s. INSAT-2 series that followed was designed based on user feedback. INSAT-2A and 2B launched in 1992 and 1993 carried VHRR payload with improved resolution of 2 km in visible and 101

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

It has been seen that even with most accurate forcing parameters and advance numerical

8 km in thermal band. The imaging capability included three modes, viz. full frame, normal mode and sector mode of 5 minutes for rapid coverage of severe weather systems. INSAT-2E launched in 1999 carried an advanced VHRR payload operating in three channels visible (2 km), thermal and water vapor (8 km). The water vapor channel is capable of giving water vapor distribution and ow patterns in the middle troposphere. Besides this, INSAT-2E also carried a CCD camera with 3 channels visible, near infrared and short wave infrared with 1 km resolution to map the vegetation cover. A geostationary meteorological satellite (METSAT) system devoted totally to meteorology was launched in 2002. It has been renamed as Kalpana-1 and is currently the operational satellite system being used by IMD. The launch of INSAT-3D in July-2013 will mark the beginning of new era in satellite observations over the Indian subcontinent region. The INSAT-3D will carry a 19-channel sounder to provide high resolution (both in time and space) observations of temperature/humidity proles and total ozone. Additionally the 6-channel imager will provide observations of earth-atmosphere variables at much higher resolution than the imagers from previous generation. Presently KALPANA and INSAT-2E satellites provide signicant services to the operational real-time weather forecasting. Some of the notable operational meteorological products derived from INSAT observations (Figure 3) are Atmospheric Motion Vector (AMV) winds, Outgoing Longwave Radiation (OLR), Quantitative Precipitation Estimate (QPE), Upper Tropospheric Humidity (UTH), Downward Shortwave (Solar) Radiation, Vegetation Index, etc. These products are extremely helpful to the meteorological analysts and forecasters for real-time forecasting. INSAT Meteorological Data Processing System, developed by Space Applications Centre (SAC, ISRO), and installed at SAC and India Meteorological Department (IMD) New Delhi, generates these products every 30 minutes. Direct measurements of radiances from KALPANA water vapor imagers are also routinely assimilated in numerical weather prediction models that produce skillful predictions of weather
Fig. 3: Products being generated operationally by Kalpana satellite

parameters at high resolution. Prediction of key weather parameters based on assimilated atmospheric conditions is provided to common

public through ISROs MOSDAC portal. Cloud images from KALPANA imagers are used for very short range prediction (called nowcasting) of convective weather. These nowcasting products are extremely helpful for ISROs satellite launch program as well as for prediction of extreme rainfall situations at very small spatial scales. Mesoscale numerical weather forecasting skill is likely to improve signicantly with the assimilation of sounding observations from INSAT-3D in near future. High resolution observations from INSAT-3D imager are also likely to improve the skill of nowcasting of convective situation.

joint collaborative project, launched in 2011 is also providing various atmosphere and ocean geophysical parameters to study to study the convective systems and their inuence on tropical weather and climate. The retrieval algorithms have been developed to provide sea surface wind speed, total precipitable MADRAS, humidity profile from SAPHIR,
Fig. 4: Sample geophysical parameters derived from Meghatropiques

radiation uxes at the top of the atmosphere from SCARAB and temperature and humidity

prole from ROSA (Figure 4). The products are being generated regularly and data is being made available through MOSDAC.

Weather Prediction
Weather forecasting is one area of operational sciences that has beneted most by advances in remote sensing technology. In early 70s, most of the observational needs for weather prediction were provided by conventional platforms like weather stations, balloons, aircrafts and ships. There were almost no observations from vast oceanic areas (particularly in the southern hemispheric oceans), deserts, the polar and the mountainous regions. Today, 95% of the data needs of operational weather prediction are fullled by satellite observations. Interestingly this chunk of weather data is only 5% of the total observations that satellites provide worldwide. It means that new and innovative methods need to be developed continuously to ingest the wealth of information from satellites that goes unutilized even today. This necessity has given birth to a new branch in operational meteorology, known by a generic name Date Assimilation. Today most of the operational weather forecasting is based on numerical models that provide weather prediction solution using the principles of uid dynamics, thermodynamics, cloud physics, radiation physics, etc. A typical weather prediction model is comprised of complex equations based on different laws of nature. These equations are solved to obtain the predictions of key weather parameters such as temperature, humidity, winds, rainfall etc. over millions of grid points within a three dimensional space. Due to the volume and complexity of these computations, developments in weather forecasting have direct link with the development of computer technology. Owing to combined impact of developments in computer as well as space-based remote sensing technologies, the skill of one week advance weather prediction in todays time is almost as skillful as the skill of two day forecast three decades ago. At Space Applications Centre (SAC), weather predictions up to 3 days in advance are provided by advanced numerical models named Weather Research and Forecasting (WRF) Model. These predictions cover entire India and adjoining sea areas at every 5 km space intervals. Observations from Indian satellites such as KALPANA, Oceansat-2, and Meghatropoiques are routinely assimilated in WRF model for high quality predictions. The forecasts of temperature, humidity, winds and rainfall, for every Indian city with population larger than 500000, are freely available to public through the ofcial website www.mosdac.gov.in. Smaller sub-sectors of large metropolitan cities like Delhi are also covered by WRF predictions. Several government and public agencies like Govt. of Gujarat, Govt. of Karnataka etc. use these weather forecasts for planning, resource management and disaster management.

103

N N R M S

B U L L E T I N

water, cloud liquid water and rainfall from

M A R C H

2 0 1 3

Meghatropiques, an ISRO-CNES

This weather prediction activity has signicant relevance to ISROs own space vehicle launch program. Weather has a large impact on many aspects of space launch that include launch operations, ground processing operations in preparation for launch, post-launch operations, and routine 24/7 weather watch and warning responsibilities. Since the successful launch of Chandrayan-I mission in 2008, the operational advisories for ISROs launch activities are largely supported by forecasts based on numerical weather prediction models like WRF.

Prediction of Tropical Cyclones

Even though the tropical cyclones are specic kind of weather systems and the problem of prediction of tropical cyclones should be covered by general weather forecasting approach, the meteorological and societal impacts of these hazardous weather systems are so strong that specic techniques and efforts are often felt necessary to address them. Operational agencies worldwide pay special attention to the monitoring and prediction of tropical cyclones, and separate models and techniques have been developed for tropical cyclone prediction. Satellite remote sensing has contributed enormously to the monitoring and prediction of these weather systems. At SAC, satellite observations are used effectively for skillful prediction of genesis, movement and landfall of tropical cyclones. For example, patterns of ocean surface winds from microwave scatterometer onboard Oceansat-2 satellite, are used to detect cyclones genesis over the warm ocean water several hours before the actual event of cyclone formation takes place. This information is crucial for several forecasting and disaster management groups to gear up for the impending hazard. Once the genesis of cyclone is conrmed, the next step is to forecast its movement. SAC developed independent techniques to predict the tracks of these systems with high skill. Several key parameters like position, intensity, and structural details of cyclone are monitored by satellite observations throughout the lifetime of a cyclone. Observations from Indian satellites Kalpana, Oceansat-2 and Meghatropiques provide valuable observations of these parameters. The present research in this area focuses on how much damage a cyclone will cause (due to winds, rainfall and storm-surge) after it makes landfall. Besides satellite based techniques, specialized numerical models designed for track and intensity prediction of cyclones are now being used for highly accurate predictions. These models not only predict the cyclone tracks accurately but also provide detailed spatial distribution of rainfall and winds across the cyclone. Such information and predictions are crucial for disaster management exercises. The land surface interface is a key component of the climate system since it provides the coupling between the atmosphere and the land biosphere and hydrology. Presently remote sensing is the most effective method for mapping land surface characteristics or energy uxes and their uncertainties globally. The land surface variables and vegetation variables, such as surface temperature, surface reectance, Normalised Difference Vegetation Index (NDVI) and Leaf Area Index (LAI) can be derived directly from the observations from INSAT-2E presently, and from INSAT-3D in near future. The regional radiant and energy ux densities can be inferred only indirectly with the aid of these land surface variables and further vegetation parameters. The monitoring of vegetation requires data of high temporal resolution. Forecasts of near surface variables (e.g. surface temperature/humidity, winds and mass/energy uxes) from numerical weather prediction models are highly sensitive to the land surface conditions, and accurate representation of land surface in model initial elds is a critical pre-requisite for a successful weather prediction. Tropical cyclones are among a few weather systems that have high societal and economic impact. Indias vast coastline is highly vulnerable to cyclone hazard. Some of the worlds deadliest cyclones have occurred over the IndiaBangladesh coastline. Satellite observations have helped a great deal in mitigating the human loss of life related to cyclone during recent decades. Observations at visible and infrared channels from INSAT satellite provide routine observations of tropical cyclones with high temporal sampling. This helps to track the position, movements and intensity of cyclones at frequent intervals, which are crucial for disaster preparedness and disaster management. Microwave observations like surface winds from Oceansat-2 scatterometer (OSCAT) are now operationally being utilized to identify the signatures of cyclogenesis. The patterns of wind circulation in OSCAT observations can be used to identify the genesis of a

development that helps forecasters, planners and disaster management agencies to gear up for an impending natural hazard. These predictions of cyclogenesis are routinely made available at ISROs MOSDAC web portal. Accurate representation of atmospheric state at the initial time (known as initial conditions) is a pre-requisite for Numerical Weather Prediction (NWP) models to generate reliable forecasts. Data assimilation (DA) systems estimate the initial condition of NWP models by making the best possible use forecast, known as background or rst guess eld. Due to complete absence of radiosondes over the oceanic area or Southern Hemisphere, measurements are not adequate to describe global distribution of humidity appropriately, thus availability of moisture observations for assimilation in NWP must rely on satellite remote sensing techniques. Radiance measurements from satellites are the major contributor to the improvement in the quality of the NWP models initial conditions and forecasts because they provide valuable information on atmospheric temperature and humidity. Direct radiance assimilation is theoretically superior to the assimilation of geophysical products because retrievals have some error in their conversion from radiances to the analysis variables such as temperature and moisture. The direct assimilation is widely used at world major operational NWP centres. Several studies have evaluated the performance of NWP models when satellite measurements were included in the assimilation process, either in the form of radiances or retrievals. Signicant positive impact on NWP model analysis has been found by assimilation of satellite radiances as well as retrievals. Recently, direct radiance assimilation capability has been developed and implemented in the WRF-Var assimilation system for some selected satellite sensors. The satellite derived radiances and products from Indian satellites such as Oceasnta-2, Meghatropiques and Kalpana as well as in-situ observations have been assimilated in Weather Research and Forecast (WRF) Model to provide local area specic weather forecast during the launch of ISRO satellites from Sriharikota Launchpad. The technology has been transferred to SHAR centre for real time forecast. Similar setup has been established at MOSDAC for All India weather forecast in 5 Km resolution and the forecast is disseminated through MOSDAC in real time. of the information provided by observations (conventional, satellite, Radar etc.) and a short-term model

Marine Lithosphere
The structural frame work of the Indian Ocean is quite complex with numerous fracture zones, abandoned spreading centers, aseismic ridges, seamounts and subduction zone. The Indian Ocean lithosphere is constantly under stress due to spreading activity south of it, resulting in fractures and intense deformation in this region. The thick sediment deposits in the Bay of Bengal mask the underlying crust and pose severe restrictions in constructing the geodynamical history of the basin. Splitting the geoid data into different wavelengths and correlating them with geology/tectonics and also their modeling have improved our understanding of the structure and dynamics of the Indian Ocean lithosphere. Advent of satellite altimetry has been a boon for offshore exploration. Space Applications Centre (ISRO), Ahmedabad has generated a very high resolution Atlas for satellite-derived free-air gravity maps over the Indian offshore using various satellite altimeter data. A number of new structures have been identied using a predictive model from satellite-derived gravity data. High resolution geoid and gravity anomaly maps (1 x 1) were generated for Indian Ocean using altimeter data from Geosat, ERS1, 2 and Topex/Poseidon satellites. Gravity anomaly maps for the Bay of Bengal and Conjugate Fracture Zones identied have been depicted in Figure 5. High resolution geoid and free-air gravity data of Bay of Bengal and Enderby Basin have also been integrated with shipborne geophysical data to investigate the early evolution of the eastern Indian Ocean. High resolution magnetic data, bathymetry 105

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

cyclone almost four days in advance. This kind of prediction provides a very early warning of cyclone

and isopach maps over the Bay of Bengal are utilized to throw light on the nature and tectonics of the BOB with particular emphasis on the 85E Ridge. The study of development of negative gravity eld of the ridge shows that the gravity anomaly of the 85E Ridge has changed from positive in the initial stages to positive at present. ONGC, GSI, INCOIS, Oil exploration companies, Maritime Boards,
Fig. 5: Gravity anomaly and residual geoid data of Bay of Bengal East Antarctica Basins. Conjugate Fracture Zones identied are marked (white line)

Indian Navy are some of the potential users of the developed methodology of deriving high resolution satellite geoid/gravity maps

and its applications in marine geosciences.

Coastal Dynamics
Regular monitoring of sediment dynamics in coastal and marine environment is essential and the ways and means by conventional point measurements using ships or boats are limited which may provide accurate concentration measurements but provide extremely poor spatial coverage that too of a particular time. Moreover, the costs of conducting such surveys are very high. Ocean colour sensors onboard satellites provide synoptic view, high repetivity and are excellent tools for understanding and modeling sediment transport. The patterns of the suspended sediment dispersals, using inhouse retrieval algorithms, bring out new information on their transport, plume dynamics, circulation patterns, currents, fronts, gyres, eddies etc. in different seasons and years. Sediment transport due to natural, human induced as well episodic events like cyclones, storm surges etc. was also studied. A methodology was developed to bring out sediment dispersal at various stages of a tide cycle exploiting the high temporal capability of OCM data as well their capability to provide improved suspended sediment
Fig. 6: Comparison of sediment plumes off Krishna-Godavari delta using numerical simulation and OCM-2 SSC images

maps. The methodology was applied for the tide-dominated regions of the Gulf of Khambhat, the Gulf of Kachchh and

the Hugli estuary to understand sediment dispersal as per the tide cycle. A technique, based on matching suspended sediment dispersion patterns, has been developed for retrieving advective vectors using sediments as a tracer from the suspended sediment maps derived from the OCM data A new approach is developed to obtain Suspended Sediment Concentration (SSC) proles from synoptic SSC derived from the OCM for a tidal environment to calculate the total sediment budget. The sediment plume dynamics from major rivers debouching into Bay of Bengal and its seasonal variations were also studied using satellite observations, numerical simulations and in situ observations. SSC derived OCM-2 data and the model simulations for region off Krishna-Godavari river mouths show a quite good match (Figure 6).

Due to the long coastline around Indian landmass, the information on implementing the CRZ Notification, monitoring CRZ violations and providing environmental clearance, zonations of the coast, demarcation of coastal vulnerability line, site selection of mangroves and shelter belts, conservation of the coral reefs,
Fig. 7: Typical mangrove community zonation and coral reef ecomorphological map using IRS

mangroves and coastal wetlands, identifying vulnerable zones due to predicted sea level

rise etc., are vital. For the last 30-35 years, studies are being carried out on various aspects of the Indian Coastal Zone. Under these studies, methodologies were developed and evolved to study the coast using satellite data, generate baseline data for the Indian coast, map and monitor the coastal landuse, coastal wetlands, critical and vital habitats (coral reefs and mangroves), shoreline changes etc. (Figure 7). These studies have become much more relevant for the country in the context of climate change and associated global warming. These studies full-ll basic data requirements of Integrated Coastal Zone Management as well as coastal disaster management, so that, suitable strategies are evolved to protect life, property and environment of the Indian coast. To predict future shorelines (hazard line), shore line changes maps and annual rate of shore line change was obtained using multidate satellite data and topographical maps. Flood height for storm surge scenarios with 25, 50 and 100 years return interval as well for sea level rise scenarios was estimated based on analysis of historic data on tides, waves, bathymetry and coastal slopes. Inundation scenarios were simulated using respective ood height for the Nellore, Paradip and Dahej coasts. In the recent period, using data of the state-of-art Indian satellites, study has been performed for the various issues pertaining to the coast. LISS-IV data has been utilized to show ecologically sensitive areas by mapping coastal landuse and landforms. Using hybrid techniques of digital and visual interpretation of Resourcesat LISS-III data, maps have been generated for mangrove community zonation and coral reef. Models for assessing health of mangroves as well coral reefs have been developed so that health bulletins may be issued at regular intervals. The methodology has been developed to assess coastal vulnerability due to sea level rise for Andhra Pradesh, Tamilnadu, Kerala and Gujarat coasts. Shoreline change mapping along the entire Indian coast using satellite data and limited eld checks has been completed. These maps depict and quantify shoreline changes as eroding/accreting/stable with reference to 1989-91 and also show status of shoreline protection measures taken by respective States.

107

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Coastal Zone Management

IRS DATA UTILIZATION FOR BIO-RESOURCES MANAGEMENT

Prakash Chauhan Space Applications Centre, (ISRO), Ahmedabad 380 115, India Email: prakash@sac.isro.gov.in

Introduction
Bio-resources are important component for progress and economic activities of a nation. Bio-resources management and utilization for human welfare is important and optimum utilization of the bio-resources is the key for sustainable development. India has rich biological diversity and one of the 12 diverse countries of the world. With only 2.5% of land area, India accounts for 7.8% of the recorded species at global level. Biological diversity means the variability among living organisms from all sources, including inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part, this includes diversity within species, between species and of ecosystems. Space based earth observation systems with their ability for a synoptic view, repetitive observations at different resolutions are an important tool for natural resource management, when compared to traditional methods of assessing natural resources. Considering Indias resource richness and the mounting pressure from the large population on the natural resources, the country needs a sustainable resources development plan. Towards this, mapping and monitoring of existing natural resources and forecasting the future scenarios are highly important. Increasing population and industrialization throughout the country and in particular along the coastal areas are adding pressure on coastal wetlands, seagrass areas and coral reefs at an alarming rate. Global warming is showing its impact, with some of the glaciers in the Indian Himalayas showing retreat. Satellite observations of land, oceans, atmosphere, and specically, during natural and human-induced hazards have become crucial for protecting the global environment, reducing disaster losses, and achieving sustainable development. Data from Indian Remote Satellites like IRS-1A/1B, IRS-1C/1D, Resourcesat-1/2, Oceansat-1/2 etc. have been extensively used for providing geo-information on various terrestrial, coastal and marine bio-resources. Remote sensing databases have been developed for large number of bio-resources management applications, such as forest cover estimation, inventory of snow and glaciers, inventory of inland and coastal wetlands, land use monitoring, Potential Fishing Zone (PFZ) forecast, coral reef and mangrove assessment etc. Table 1 shows the list of satellites/sensors mainly used in last 25 years to generate databases on the bio-resources management for terrestrial, coastal and marine bio-resources. Historically, application of remote sensing data to solve common mans problems in India was started with an example for bio-resource management experiment of detection of coconut root wilt disease from colour infrared imageries in 1970. Since then Indian remote sensing applications have matured to the present status of operational use in bio-resources inventory generation and management. The main thrust of environmental and bio-resources assessment studies aims at

Satellites (year) IRS-1A/1B (1988, 1991)

Sensor LISS I

Spectral bands(m) 0.450.52 (B) 0.52-0.59(G) 0.620.68 (R) 0.770.86 (NIR) Same as LISS-I Same as LISS-I 0.520.59 (G), 0.62-0.68 (R) 0.770.86 (NIR) 1.55-1.70 (SWIR) 0.62-0.68 (R) 0.770.86 (NIR) 0.50-0.75 0.7550.768 (4 bands) 0.4081.010(13 bands) 1.6 (1 band) 0.620.68 (R) 0.770.86 (NIR) 1.551.70 (SWIR) 0.4020.885 (8 bands) 6.6, 10.65, 18, 21 GHz (V&H) 0.520.59 (G) 0.620.68 (R) 0.770.86 (NIR) 0.520.59 (G), 0.620.68 (R) 0.770.86 (NIR) 1.551.70 (SWIR) 0.520.59 (G), 0.62-0.68(R) 0.770.86 (NIR) 1.551.70 (SWIR) 13.4 GHz 0.4020.885 (8 bands) 0.520.59 (G), 0.62-0.68(R) 0.770.86 (NIR) 1.551.70 (SWIR) 0.520.59 (G), 0.620.68 (R) 0.770.86 (NIR) 0.770.86 (NIR) 1.551.70 (SWIR 0.520.59 (G) 0.620.68 (R) 0.770.86 (NIR) 5.3 GHz, HH, VV, HV

Spatial res. (m) 72.5

Swath (km) 148

Radiometric res. (bits) 7

Repeat cycle (days) 22

LISS-II IRS-P2 (1994) IRS-1C/1D (1995 / 97) LISS-II LISS-III

36.25 36.25 23.5

74 74 141

7 7

22 24

7 70.5 (SWIR) 188 5.8 1570 1400 520 520 520 640 188 148 810 70 195 200 192 810 7 6 16 16 16 7

24 24 (5) 24 (5) 24 24 24 5

WiFS PAN IRS-P3 (1996) MOS-A MOS-B MOS-C WiFS

IRS-P4 (1999)

OCM MSMR

360 236 150, 75, 50 & 50 km respectively 5.8

1420 1360

12 -

2 2

IRS-P6 (2003)

LISS-IV

70

10 (7)

24 (5)

LISS-III

23.5

141

24

AWiFS

56

737

10

24 (5)

Oceansat-2 (2009) Resourcesat-2 (2011)

Scatterometer OCM AWiFS

50 km 360 236 56

LISS III

23.5

23-70

10-12 7-10 7-10

24

LISS IV

5.8

RISAT-1 (2012)

SAR

350

25(MRS ) 13(CRS )

B= Blue, G= Green, R= Red, NIR= Near Infrared, SWIR= Shortwave Infrared 109

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Table 1: Major specications of IRS series satellite & sensors system used of bio-resources assessment

monitoring and understanding of structure, composition and function of different vulnerable ecosystems to regulate the impact of developmental activities and sustain the delivery of natural ecosystem functions and services.

Forest Studies
One of the important missions has been the rst national level remote sensing-based mapping of forest cover of the country in 1983. Since then, the Forest Survey of India (FSI) is carrying out biennial forest cover mapping. FSI assesses forest cover of the country by interpretation of remote sensing satellite data. The major activities of FSI include forest cover assessment, inventory of forest areas, thematic maps, inventory of trees outside forests (rural & urban), inventory data processing, methodology design, training and extension, projects and consultancies. FSI publishes all the analysed data and results in a biennial report called State of Forest Report (SFR). Indian SFR 2011 report has been the most recent update on the forest cover of India and FSI has used IRS-P6 LISS III data at 23 m spatial resolution to generate this report. Table 2 shows the recent status on forest cover of India. Table 2: Forest and Tree cover of India in 2011 Class Forest Cover a) Very Dense Forest b) Moderately Dense Forest c) Open Forest Total Forest Cover* Tree Cover Total Forest and Tree Cover Scrub Non-forest Total Geographical Area
(*Includes 4662 km2 area under mangroves)

Area (km2) 83,471 320,736 287,820 692,027 90,844 7,82,871 42,177 2,553,059 3,287,263

% of Geographical Area 2.54 9.76 8.75 21.05 2.76 23.81 1.28 77.67 100.00

The advent of high resolution IRS-P6 Resourcesat has enhanced the capacity to prepare forest type and density maps and generation of forest-working plans. Forestation and deforestation could also be assessed using multi-temporal satellite data. Biome level vegetation characterization for entire India using multi-temporal IRS WiFS data and broad level vegetation type mapping for the entire country on 1: 50k scale using IRS LISS III data have been carried out. The information generated by the aforesaid programmes is extensively used in conjunction with ground based related information in geospatial domain for spatial explicit quantication and process understanding related to biodiversity assessment, landscape change and simulation, carbon sequestration, hydrology, generic ecosystem patterns, EIA studies and regional climate models. One of the important studies undertaken in this regard is the collaborative effort on landscape level characterization of biodiversity using remote sensing and GIS.

Coastal Zone Studies

Coastal zones are one of the most ecological sensitive areas and since a large population resides in coastal zones they are under continuous stress. Mangrove forests, coral reefs and wetlands are critical habitats of the coastal zone. Information required for the purpose of coral reefs includes spatial distribution of the reefs, vegetation cover, reef zones and reef morphology, biodiversity of fauna and condition assessment. Earth Observation (EO) data have been used to prepare maps showing the extent and condition of coral reefs and the extent, density, condition and diversity of the mangroves. Land use/land cover dynamics and its prediction, retrieval of forest parameters using high-spatial resolution and hyperspectral data, species identication, biodiversity monitoring and change modelling, vegetation stress analysis, disease detection, forest ecosystem responses to climate change and anthropogenic impacts and ecological foot

are a few important areas which have been studied using remote sensing data from IRS series of satellites. Remote sensing data from Indian missions also forms an important component for Coastal Regulation Zone (CRZ) implementation in the country. Coastal landuse maps of entire Indian coast has been These maps show Ecologically Sensitive Areas (ESAs) and high tide and low tide lines.
Fig. 1: Shoreline changes along the Alibag coastal region in Maharastra LISS-III data of 2004.

These maps are utilised for CRZ monitoring and implementation. Shoreline changes and coastal process studies have been done using high resolution LISS-III and LISS- IV data. Figure 1 shows the shore line change along the Alibag coastal stretch in Maharastra.

Mangroves Studies
Remote sensing data derived from different satellite systems have provided information about the areal extent, conditions and boundary of coastal wetlands. Further, multi-dated satellite data has been effectively used to nd out the changes in the extent of mangroves. IRS (especially IRS-1A, 1B, 1C) has
Fig. 2: Examples of deltaic mangroves (IRS-1D LISS III image on February 28, 1999) part of Brahmani-Baitrani deltaic complex showing mangroves of Bhitarkanika region

been quite extensively used for monitoring and change detection of some critical habitats like mangroves and coral reefs in Indian

coastal regions. Some of the studies also focus on zoning the dominant mangrove communities using higher resolution data from IRS-1C/1D LISS III and PAN data. IRS-1D LISS III images helped extensively in identication of different mangroves communities such as Excoecaria-Avicennia, Avicennia and Avicennia-Sonneratia community in Coringa creak and Bi Bhitarkanika region (Figure 2). First spatial Table 3: Mangrove area (km2) in different states as monitored using satellite data Region Gujarat Maharashtra Goa Karnataka Tamil Nadu Andhra Pradesh Orissa West Bengal Andaman & Nicobar Total Area (1986) 767.0 124.0 5.5 11.30 30.0 322.0 203.0 2067.0 922.0 4451.8 Area (1990-93) 1014.6 222.6 6.7 8.7 23.6 380.0 187.0 1838.0 749.9 4431.1 Area (2005-07) 890.7 270.9 34.6 6.1 57.3 351.3 221.1 2529.3 588.3 4949.6 111

N N R M S

B U L L E T I N

prepared using LISS-IV data on 1:25k scale.

M A R C H

2 0 1 3

printing analysis for sustainable development

inventory of Mangroves have been made by Space Applications Centre (SAC), Ahmedabad using satellite from IRS instruments for the period of 1989-91 time frame. Subsequently similar inventory of mangroves on 1:25k scale was done for entire Indian coast during 2005-07. EO data have also been used to prepare maps showing the extent and conditions of the mangroves at community level. Table 3 shows the mangrove area in different states monitored using primarily IRS series data sets over the time period of 1986 to 2007. The mangrove inventory shows an overall increase in mangrove areas by ~11%.

Coral Reefs
Coral reefs offer benet to mankind in terms of food production, as tourism, recreation and shoreline protection. The world conservation strategy identied coral reef as one of the essential life supporting system necessary for food production, health and other aspects of human survival and sustainable development. Corals act as a barrier against wave action along coastal areas thus preventing coastal erosion. In addition, coral reefs protect mangroves and sea grass beds in certain areas, which are the breeding and nursing grounds of various economically important fauna. Because of the increasing human population along the coastal area, anthropogenic impacts on the coastal zone have become severe over the past few decades. Remote sensing technologies have great potential in providing information about coral reefs for their monitoring, change detection and delineation. The latest Indian satellites IRS-1C, 1D and IRS-P4 with their improved spatial resolution (PAN-5.8 m, LISS-II-23.6 m, WiFS-188 m) extended spectral resolution (inclusion of middle infra-red band in LISS-III) and increased repitivity (5 days for WiFS data) has opened up new vista for application in coastal zone. Preliminary analysis of IRS-1C, 1D indicates that coral reef zonation, identication of trees and shrub mangroves, mudats, beach dune vegetation and saline area etc. possible now. The PAN data combined with LISS-III are extremely useful in providing detailed spatial information about reclamation, construction activity and ecological sensitive areas which are vital for coastal zone regulatory activities. Information available from merged PAN and LISS-III data about coral reef zonation especially for atolls, patch reef are valuable for coral reef conservation plan. Coral reef mapping for the entire Indian coast has been carried out using IRS data of 2004-07 . As per the inventory the coral reef area for India is around 2383.9 km2. Table 4 shows the geographical distribution of area under coral reefs for the entire Indian Territory. Table 4: Geographical distribution of coral reefs in India Region Gulf of Kutch Malwan, Maharastra Lakshadweep Gulf of Mannar Andaman & Nicobar Area (Km2) 352.5 0.28 933.7 75.9 1021.5

IRS LISS-III data visually analyzed to map coral reef features such as type of coral reefs, mud and sand over reef, coralline shelf, reef vegetation, sand beach vegetation, algae, sea grass, sea weeds, type of lagoonal bottom, live corals as well as high and low water lines. IRS LISS-III data and PAN merged products was also evaluated for coral reef zonation. Figure 3 shows example of moderate resolution (23.5 m) LISS-III image data and its merged image with PAN at high resolution (5.8 m) in same area. Sedimentation of reefs reduces live corals and species diversity and sh biomass is affected. Lakshadweep islands are mostly atolls with few coral heads, a platform reef and sand cays. Coral reefs of Gulf of Kutch and few reefs of Andaman and Nicobar islands are in degraded condition as indicated by mud deposition.

Glaciers are normally described as mass of ice slowly moving down a gradient. A glacier consists of ice crystal, water and rock debris. Out of these, ice is an essential part of the glacier. Satellite remote sensing data has been used to map and monitor the
(a) (b)

Himalayan glaciers. Glacier inventory of Indian Himalaya was completed at 1:250k scale using IRS LISS-II data. 1702 glaciers covering an area 23,300 sq km were mapped. Glacier inventory at 1:50k scale was carried out in the Indus, Ganga and Brahamputra basins. Initial studies on monitoring of glacial retreat and estimation of loss in glacial area was carried out for two glaciers viz. Samudra Tapu and Parbati glaciers in Himachal Pradesh. Satellite images showed that these glaciers are presently retreating at the rate of 26 m and 38 m per year, respectively (from base year 1962). A model based on accumulation area ratio has been developed to estimate glacial mass balance. The model suggests an annul loss of 84 cm of glacial ice in Baspa basin.
B U L L E T I N

Fig. 3: (a) Paga coral reef in Gulf of Kachchh images using IRS-P6 LISS-IV data on Match 16, 2005 showing various geo-morphological features. (b) coral reef eco-morphological zonation map at 1:25k scale for Kavaratii island in Lakshadweep

(a)

(b)

Fig. 4: (a) Various glacial features shown on Resourcesat-2 LISS-IV data. (b) Retreat of Gangotri glacier as seen from different IRS satellite data sets during the year 1962 to 2006. Data for 1962 has been taken from Survey of India toposheetLakshadweep

This study is now expanded to 10 sub-basins in the Indian Himalaya. A recent study done by SAC, Ahmedabad show that 75% of the Himalayan glaciers have retreated over the 15 year period while 18% advanced and rest showed no change. Seasonal snow cover monitoring in the Baspa river basin suggests a large retreat in December and average stream runoff of the month has increased by 75%. The observations such as glacial retreat, negative mass balance, early melting of seasonal snow cover and wintertime increase in stream runoff suggest an inuence of climate change on the Himalayan cryosphere. Figure 4 shows important glacial features marked on satellite imagery and the retreat of Gangotri Glacier as observed from 1962 to the year 2007 using the satellite data. Similarly, change in snow cover is also expected as an indicator of climate change. Snow cover monitoring algorithm using Normalized Difference Snow Index (NSDI) technique was used to monitor seasonal snow cover in 28 sub-basins in the Indus and Ganga river basins using AWiFS data from Resourcesat-1. Use of the SWIR band of AWiFS data has provided better snow and cloud discrimination.

Marine Living Resources

Marine and freshwater fish stocks are an important high-protein food source for humankind. With a large coastline, India is one of the largest sh producing countries across the world. Fisheries play an important role in the nations economy, providing full-time or part-time employment to 14.66 million people in India. But, the shing industry is coming under increasing pressure from a range of factors related to the growth in human population, including over-fishing, global climate change, pollution, and habitat degradation. The result is that some 80% of the world sh stocks are either now fully exploited or overexploited. 113

N N R M S

M A R C H

2 0 1 3

Glacier Studies

Ocean color remote sensing started in India with the successful attempts by scientists of FAO/UNDP Pelagic Fisheries Project. They succeeded in direct spotting of sh schools and their quantication by aerial surveys and vessel surveys during the years 1972-74. The next big leap in the remote sensing of fisheries was the mapping of chlorophyll distribution in the Cochin coastal area by scientists from National Remote Sensing Agency (NRSA) and Central Marine Fisheries Research Institute (CMFRI), making the way for modern marine remote sensing. During 1980-84 three major organizations, the SAC, CMFRI and Fishery Survey of India (FSI) joined together
Fig. 5: Chlorophyll distribution in Indian Exclusive Economic Zone (EEZ) using OCM data

to collaborate in the Joint Experiment Programme (JEP) for marine sheries. The main objectives of the experiments were to identify suitable sensor parameters such as optimum spectral bands, sensitivity, and special resolution in the context of the Indian Remote Sensing Satellite Utilization Programme. Remote sensing applications in sheries management has many facets of applications which includes its use in sh harvesting, stock assessment, marine aquaculture, marine recourse management, sheries productivity assessment, study on impact of climate change and ecosystem variability on sheries. Indian satellites contribution to sheries has many success stories to tell.

Fig. 6: Oceansat-1 OCM derived Chlorophyll-a of November 17, 1999 along Goa coast forecasting potential shing zone along the meander of chlorophyll marked by arrows. Fish Catch feedback from Fisheries survey of India (FSI) conrmed the forecast results

Use of Chlorophyll-a data from IRS-P3 MOS-B, Oceansat-1 OCM and more recently Oceansat-2 OCM along with Sea Surface Temperature (SST) data has been used for

demarcation of Potential Fishing Zone (PFZ) through earth observation is currently an operational technology, which is based on the principle of identication of feeding grounds where sh tend to accumulate. The technique developed for the PFZ forecast (up to 23 days in advance), has been validated with a number of ship campaigns in the Indian waters. Results have shown 7090% success in PFZ identication. PFZ advisory is regularly given by Indian National Centre for Ocean Information Services (INCOIS), Hyderabad to local shermen, beneting large shermen community of Indian maritime states. ISRO is planning advanced ocean colour sensor series of Oceansat-3 and ocean colour from Geo-Stationary platforms. Figure 5 shows the chlorophyll distribution in Indian Exclusive Economic Zone (EEZ). Figure 6 shows an example of a chlorophyll map used for detection of PFZ areas along with results of experimental shing.

scientic community for other applications like locating tuna forage ground, explaining the variability in Indian oil sardine stock, shrimp culture in India, bloom monitoring in the sea, new production studies.

Future Perspective
Earth Observation program in India is primarily application driven. A large infrastructure for ground segment has been established in the country to deliver the benets of space based EO services. with multi-institutional framework and peoples participation in decision making process. This would turn the direction of working from being a data provider to service provider by giving end-to-end solutions. As part of this, food security, water security, environmental monitoring and infrastructure development are going to be the mainstay of applications with a focus on rural and community development. On the other hand, the ecosystem responses and disaster monitoring and mitigation, and climate change studies would also stand as important activities as they would ultimately impact the overall development. The need of the hour is also to work on process models considering Earth as an integrated system. The concept of earth system science needs to be strengthened and space based observations needs to be integrated with numerical process models. In order to cater the need of societal applications and earth system science programs there is requirement for development of new sensors as well as to have continuity of the existing missions. Some the important future EO missions of ISRO are INSAT-3D and GISAT. Data from these new missions and the recently launched SARAL/ Altika would ll the gaps in terms of spectral, spatial and temporal resolutions in both optical and microwave regions. The space-borne microwave and hyperspectral sensors would form the important basis in ecosystem quantication, retrieval of geophysical and biophysical parameters. In order to realize the above missions and derive maximum benet from their use, a number of areas, especially related to ground segment need to be addressed simultaneously. The most important among these is to generate a large number of bio-geo-physical products on near-real time basis and disseminate to users through easily accessible means However, future EO program will strongly depend on providing Integrating EO products and services

115

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Indian remote sensing data with other international mission data has also been used by

REMOtE SENSING APPLICAtIONS USING IRS DAtA IN GEOSCIENCES

Vinod Kumar K, John Mathew, Tapas R. Martha, Arindam Guha, Babu Govindha Raj K, Ritwik Majumdar and Priyom Roy National Remote Sensing Centre, Hyderabad 500 037, India Email: vinodkumar_k@nrsc.gov.in

Introduction
Remote sensing is one of the important tools being used for major exploration programme in the country. In the past decade, the eld of geosciences has made a signicant progress in utilizing remote sensing technology for many of the applied geological problems. Geological remote sensing has evolved from the earlier days of photo-geological interpretation to the present day use of hyperspectral remote sensing for identication of minerals. The availability of satellite data in different regions of electromagnetic spectrum has contributed signicantly for these developments. Remote sensing of geophysical parameters like gravity, magnetism and electromagnetism provide valuable information about the subsurface geology on the basis of anomalies resulting from the physical properties of the subsurface rocks. The sensors onboard the Indian Remote sensing (IRS) satellites have contributed immensely in various domains of geology ranging from direct applications like geomorphological/lithological/ structural mapping to applied aspects like geo-exploration, geo-environmental and geo-hazard studies. These applications which initiated subsequently to the successful launch IRS-1A (with LISS-I and LISS-II sensors having resolutions of 72.5 and 36.25 m respectively) in 1988 have matured since then with the availability of higher spatial and spectral resolution and also imaging capabilities beyond VIS/NIR/SWIR to TIR and Microwave domains. The imaging capabilities have achieved milestone with the operationalization of RISAT-1 in 2012. The sensors from the recent IRS series like AWiFS, LISS-III, LISS-IV and Panchromatic (Cartosat-1 & 2) have contributed signicantly in various geosciences themes. This article tries to highlight a few case studies which have used the IRS data in some of the geo-scientic applications.

Literature Review
There are innumerable studies utilizing IRS data various geological applications, however the description below contains only some of these studies. A few of the initial examples using IRS data in geosciences have been for ground water exploration and mineral exploration studies like the national hydro-geomorphic mapping on 1:250K, mineral exploration project (Project Vasundhara) etc. Hydrogeomorphological mapping involves identifying the landforms and evaluating their ground water prospects. Ravindran and Jeyaram (1997) used IRS-1B LISS-II data for hydrogeomorphic mapping in the Baran district of Rajasthan and identied prospective areas for ground water development. Saraf and Choudhury (1998) demonstrated integrating IRS LISS-II data and terrain data for identication of sites of ground water exploration and articial recharge. Sahai et al., (1991), Krishnamurthy and Srinivas (1995), Krishnamurthy et al., (1996) also utilized remote sensing data

the existing geological and structural maps in the inaccessible terrains like the Himalaya. Philip et al., (2003) used IRS data for discriminating the lithological units of the Nidar Ophiolite Complex of the Indus suture zone and also for delineating the landforms of the area. The application of remote sensing data in mineral exploration is related to identifying and extracting the controls of mineralization like structural, lithological or geomorphological and then integrating with geochemical or geophysical data for targeting the prospective locations for detailed characteristics of various mineral or mineral groups with that available from the remote sensing data for mineral identication or for quantitative analysis. Ramadass et al., (2006) integrated IRS LISS-III data and gravity data for delineating prospective kimberlite zones in eastern Dharwar craton. Srivastav et al., (2000) integrated IRS LISS-II data and other remote sensing data with geophysical and geochemical data for locating favourable zones of leadzinccopper deposits in the Rajpura-Dariba area of Rajasthan. IRS data has been widely been used in different applications related to geo-hazards (mainly earthquakes, landslides and volcanism; in Indian context). Studies like disaster risk identication and evaluations are commonly utilized remote sensing data. It has also been applied for damage assessment and relief/rehabilitation measures in the post-disaster phase. Remote sensing data has also been used by scientists to map the active faults for evaluating the associated neo-tectonic activity. Kumar et al., (2006) assessed the damage of the 2005 Kashmir earth quake using IRS Cartosat-I, AWiFS, LISS-III and LISS-IV data. They identied a series of landslides associated with the MBT and Jhelum Fault near Muzzafarabad and along the MBT in the Uri region, triggered by the earthquake and also mapped the damage to property and infrastructure associated with it. Sitharam et al., (2006) evaluated the seismic hazard of Bangalore using remote sensing and seismotectonic parameters. They had used IRS data along with data from other sources for extracting the geological structures. Another widely used domain of remote sensing data application is in landslide hazard related studies for various tasks like preparation of landslide inventory (using manual or semi-automatic procedures) and for mapping of various pre-conditioning factors of landslides towards landslide susceptibility mapping through knowledge driven or data driven methods. Martha et al., (2010) proposed object oriented approach for semi-automatic detection of landslides in the Himalaya. Mathew et al., (2007, 2009) used IRS multispectral and PAN data for identifying the landslide susceptible areas in the Garhwal Himalaya using statistical approach. Volcanic activity is another geological hazard where remote sensing may be applied for various aspects like monitoring the temperature variations or for mapping the extent of volcanic ash deposits and lava ow extent, post disaster (Kumar et al., 2006a). Geo-environmental application of remote sensing is related to the monitoring of land-use changes associated with mining, acid mine drainage analysis using hyperspectral remote sensing, land subsidence mapping using microwave interferometric studies, underground coal re extent mapping using thermal remote sensing data, glacial retreat studies etc. Prakash and Gupta (1998) studied the land use dynamics related to coal mining in Jharia coaleld using IRS LISS-II and other data. Babu Govinda Raj et al., (2013) used IRS data alongwith other data for data for identifying the glacial lakes and the associated threat of glacial lake outburst ood in Sikkim Himalaya. Remote sensing data has also been in geo-engineering projects. Pramod et al., (2010) used IRS data for identifying and characterizing potential hydropower sites in the upper Alakananda and Bhagirathi valleys of Uttarakahand. exploration. With the availability of hyperspectral data, this has advanced to matching the spectral

117

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

for ground water exploration studies. Many studies also demonstrate the utility of IRS data in rening

Use of IRS Data in Geological Applications: Case Studies

National Geomorphology and Lineament Mapping Geomorphology is a complex subject with multiple approaches to analysis and multiple scales of mapping. The nature of the geomorphic unit is controlled by the model of analysis chosen and the scale of mapping required. The basic geomorphic unit is not a single feature or entity, but must be carefully selected to be essentially homogeneous and indivisible at the scale chosen. Although all generally agree that the basic geomorphic unit should be a homogeneous entity, it can be dened in terms of genetic or structural pattern, the approach followed by the IGU and most European geomorphologists, or in terms of location and dimensions of geometric elements, along the lines of the British system. In India, very few attempts have been made to bring out a regional classification system addressing all the
Fig. 1: Geomorphology and lineament map of Goa state

aspect of the landform evolution. National

geomorphological and lineament mapping is one major effort, using the capability of remote sensing data like IRS, for mapping the diversity of landform in the country IRS LISS-III data is operationally used to create the national data base for geomorphology and lineament in association with Geological Survey of India. An inventory 417 landform units is being made covering the diverse geological / geomorphological provinces in the country. The landforms are grouped based on their genetic origin into various classes with the emphasis on morphography, morphogenesis & morphostructure. These maps are generated on 1:50K to be integrated with other geoscientic data base like lithology, geochemical and geophysical data for applied geological applications (Figure 1). IRS Data in Geological Structural Studies
Fig. 2: Two different generations of folds are shown; axial planes of fold are marked as F2 and F3 respectively. F2 is regional plunging fold; where F3 is preserved in the limb of F2 fold. Isoclinal folds are also shown in 2.A; which are developed during F2 folding

IRS P6 LISS IV data of the BayanaLalsot Basin, Rajasthan is analysed to infer the

stages as evident from the distinct outcrop patter ns and their association. The interpretation of closure of folds in plan view indicates different types of folds with their fold axes plunging towards different direction, are preserved in Alwar rocks of this basin. The trend or plunge of major faults are trending along NE-SW and NW-SE directions (Figures 2 & 3). These faults have distinct strike slip component in plan. These geological structures are favourable loci for
Fig. 3: The high pass ltered image IRS LISS-IV shows two faults (shown by red dash line) displaced the limb of large fold (latest regional fold)

mineralization. IRS Data in Geological Studies of Potential Hydropower Sites in Myanmar IRS data has got applications in geoengineering related projects like hydro-power potential identication and characterization, tunnel alignment projects, road alignment along hilly slopes etc,. The application in hydropower feasibility projects vary from evaluation of topographic head using Digital Elevation Models (DEMs) derived from satellite stereo data (e.g. Cartosat-I stereo data), discharge estimation at the selected potential Hydro Electric Power (HEP) sites using measured discharge values modeled using catchment characteristics derived from

Fig. 4: Tamanthi area: highlight of the study

high resolution satellite data, evaluation site of suitability based on geological structures and slope stability, evaluation of reservoir tightness etc. ISRO has carried out Remote Sensing based Geological Studies around the Tamanthi and Shwezaye HEP projects in Myanmar for NHPC Ltd. The objectives of the study comprised of lithological, structural and geomorphological interpretation from IRS data pertaining to the proposed HEP sites and the surrounding areas falling in the Chindwin valley of Myanmar (Figure 4). The structural, lithological and geomorphological mapping on 1:100K scale were carried out for the upstream catchment area of Chindwin river from the proposed dam sites at Tazon (Tamanthi) and Shwezaye sites and on 1:10K scale in and around the sites. The study used IRS Resourcesat-I AWiFS, LISS III, LISS IV and Cartosat I PAN data. Some of the structural features interpreted from the IRS data have been highlighted in the gure, along with their ground expressions. IRS Data in Landslide Incidence Mapping and Landslide Volume Estimation Landslide incidence mapping or the annual inventory of landslide is one important application 119

N N R M S

B U L L E T I N

fold axis is along NE-SW direction. The

M A R C H

2 0 1 3

presence of folds of different deformational

using IRS data sets. The incidence of landslide can be mapped from LISS-IV satellite data supported by Digital Surface Model (DSM) derived from Cartosat-1 data sets. Landslide inventory is being routinely carried out using IRS data after any major event of landslide. Landslide inventory map of Okhimath prepared from Resourcesat-2 LISS IV Mx image (20-sept-2012) and Cartosat-1 DEM through a semi-automatic approach is shown in Figure 5. In a study in the Garhwal Himalayas of Uttarakhand it was shown how pre- and post-landslide DEM can be used to estimate the volume of Salna landslide (July 2007), without using additional ground information in the form of Ground Control Points (GCPs). In this study the volume of the landslide was estimated as 1.43 million cubic meters (Figure 6). IRS data in Landslide Hazard Zonation (LHZ) Mapping along the Pilgrimage Route Corridors of Uttarakhand and Himachal Himalaya
Fig. 5. Landslide inventory map of Okhimath prepared from Resourcesat-2 LISS IV Mx image (20-sept-2012)

National Remote Sensing Centre (NRSC), carried out Landslide Hazard Zonation (LHZ) Mapping along the corridors of the important pilgrimage/tourist routes in Uttarakhand and Himachal Himalaya in 20002001. The project covered a total length of about 2000 km and utilized merged IRS 1D PAN and LISS III data of 1998-2000, on 1:25K scale for mapping of various contributory factors of landslides as well as the landslide

Cartosat-1 Orthoimage of Apr6 2006

Cartosat-1 Orthoimage 16Dec 2007

Volume estimation of Salna landslide

incidences. Lithology, lineaments, slope-dip relation, geomorphology, slope, slope aspect and morphology, land use/land cover, soil

Fig. 6: Estimating volume of Salna landslide

depth, soil texture, weathering, drainage etc were mapped on 1:25K scale along the route corridors (Figure 7). These parameters were interpreted using satellite data and were ground veried. The corridor width was limited from ridge crest to ridge crest on either side of the main valley (limited to maximum of 5 km on either sides) along which the roads are aligned. The landslide incidences were also mapped since understanding of the existing landslide scenario is important in quantifying the relative contribution of various causative factors towards slope destabilization. The landslide polygons were given the attribute as old or active based on image interpretation and subsequent ground validation. The corridors studied in Uttarakhand were: 1) Rishikesh-Rudraprayag-Chamoli-Badrinath, 2) RudraprayagOkhimath-Kedarnath, 3) Chamoli-Okhimath, 4) Rishikesh-Uttarkashi-Gangotri-Gaumukh and 5) Pithoragarh-Khela-Malpa.

1) Dalhousie-Chamba-Brahmaur, 2) ShimlaBilaspur-Kulu-Manali and 3) Shimla-RampurSarhan-Sumdo. The database was integrated using Analytical Hierarchical Process (AHP) modelling which involved pair-wise comparison of to derive the weights of parameters and their classes. Spatial integration of the weighted themes and classes resulted in the cumulative relative contribution of all the factors and was segmented to generate LHZ classes as Very Low/Moderate/High and Very High hazard classes. Data in Glacial Landform Mapping and Glacial Retreat Studies Glaciers are highly significant components of the global climate system, affecting atmospheric and oceanic circulation systems, the amount of light reflected off the planet and the hydrological cycle. Reconstructing former glaciers and their dynamics provide insights into the behaviour of the present-day ice masses and potentially how they will respond to, and affect, future climate change. An understanding of
Fig. 7: Landslide Hazard Zonation (LHZ) map around Chamoli, Uttarakhand

Pleistocene to Recent glacial episodes will also help us to model how future glacial episodes could affect our landscape. IRS data has been used to understand the various glacial landform morphology and glacial retreat as shown in Figure 8.

Discussion and Conclusion

The case studies given here are only indicative applications from a wide realm of potential use of IRS data in geological sciences. These
Fig. 8: Glacial landform map and retreat map of Milam Glacier using IRS data sets (2.5 km retreat is seen)

demonstrate that IRS data have contributed in various programmes relevant to the natural resources management and development of

the country. The spectrum of IRS sensors with large swath and moderate-low spatial resolution to those with narrow swath and very high spatial resolution had been the mainstay of remote sensing based geological applications in India during the past and at present. The evolution from LISS-I to the 121

N N R M S

B U L L E T I N

various contributory factors and their classes

M A R C H

2 0 1 3

In Himachal Pradesh the corridors included:

latest LISS-IV sensor in optical domain and the introduction of microwave and hyperspectral imaging capabilities have provided geoscientists with many options to utilize the remote sensing data to study various applied or direct aspects. Development of sensors with high-moderate resolution, thermal and hyperspectral imaging capabilities, alongwith high spatial resolution optical and microwave data in the near future will further enhance these options.

References
Govindha Raj, B.K., Remya, S.N. and Kumar K.V. (2013). Remote sensing-based hazard assessment of glacial lakes in Sikkim Himalaya. Current Science, 104(3), 359-364. Krishnamurthy, J. and Srinivas, G. (1995). Role of geological and geomorphological factors in ground water exploration: A study using IRS LISS data. International Journal of Remote Sensing, 16(14), 2595-2618. Krishnamurthy, J., Kumar, N.V., Jayaraman, V. and Manivel, M. (1996). An approach to demarcate ground water potential zones through remote sensing and a geographical information system. International Journal of Remote Sensing, 17 (10), 1867-1884. Kumar, K.V., Martha, T.R,. and Roy, P.S., (2006). Mapping damage in the Jammu and Kashmir caused by 8 October 2005 Mw 7.3 earthquake from the Cartosat1 and Resourcesat1 imagery. International Journal of Remote Sensing, 27(20), 4449-4459. Kumar, K.V., Martha, T.R., and Roy, P.S., (2006a). Detection of volcanic eruption in Barren Island using IRS P6 A WiFS data. Current Science, 91(6), 752. Martha, T.R., Kerle, N., Jetten, V., van Westen, C.J. and Kumar K.V. (2010). Characterising spectral, spatial and morphometric properties of landslides for semi-automatic detection using object-oriented methods. Geomorphology, 116 (1), 24-36. Mathew, J., Jha, V.K. and Rawat, G.S. (2007). Application of binary logistic regression analysis and its validation for landslide susceptibility mapping in part of Garhwal Himalaya, India. International Journal of Remote Sensing, 28(10), 2257-2275. Mathew, J., Jha, V.K., and Rawat, G.S., (2009). Landslide susceptibility zonation mapping and its validation in part of Garhwal Lesser Himalaya, India, using binary logistic regression analysis and receiver operating characteristic curve method. Landslides, 6 (1), 17-26 Philip, G., Ravindran, K.V. and Mathew, J. (2003). Mapping the Nidar ophiolite complex of the Indus suture zone, Northwestern-Trans Himalaya using IRS-1C/1D data. International Journal of Remote Sensing, 24 (24), 4979-4994. Prakash, A. and Gupta, R.P. (1998). Landuse mapping and change detection in a coal mining area- a case study in the Jharia coaleld, India. International Journal of Remote Sensing, 19 (3), 391-410. Pramod Kumar, Mathew, J., Giribabu, D., Sharma, K.P. and Krishnamurthy, Y.V.N. (2010). Potential hydropower sites investigation (pre-feasibility level) in parts of Alakananda and Bhagirathi valleys, Uttarakhand using geospatial techniques. Technical report, RRSC-D/ISRO, 75p. Ramadass, G., Himabindu, D. and Veeraiah, B. (2006). Morphostructural prognostication of kimberlites in parts of eastern dharwar craton: Inferences from remote sensing and gravity signatures. Journal of the Indian Society of Remote Sensing, 34 (2), 111-121. Ravindran, K.V. and Jeyaram, A. (1997). Groundwater prospects of Shahbad tehsil, Baran district, Eastern Rajasthan: a remote sensing approach. Journal of the Indian Society of Remote Sensing, 25 (4), 239-246. Sahai, B., Bhattacharya, A. and Hedge, V.S. (1991). IRS-1A application for groundwater targeting. Current Science, 61, 3-4. Saraf, A.K. and Choudhury, P.R. (1998). Integrated remote sensing and GIS for groundwater exploration and identication of articial recharge sites. International Journal of Remote Sensing, 19 (10), 1825-1841. Sitharam, T.G., Anbazhagan, A. and Ganesha Raj, K. (2006). Use of remote sensing and seismotectonic parameters for seismic hazard analysis of Bangalore. Natural Hazards and Earth Systems Sciences, 6, 927-939. Srivastav, S.K., Bhattacharya, A., Kamaraju, M.V.V., Reddy, GS., Shrimal, A.K., Mehta D.S., F. K. List Burger, H. (2000). Remote sensing and GIS for locating favourable zones of lead-zinc-copper mineralization in Rajpura-Dariba area, Rajasthan, India. International Journal of Remote Sensing, 21(17), 3253-3267.

ACHIEvEMENtS AND APPLICAtIONS FOR URbAN PLANNING USING DAtA FROM IRS SAtELLItES
Venugopal Rao, Kamini J, Ramesh B and Raghavaswamy V National Remote Sensing Centre, Hyderabad - 500 037, India Email: venu_koppaka@nrsc.gov.in

Introduction
The urban system is a complex geographical synthesis of human and nature. It is also a sign of civilization, national politics, economics and culture. The level of urbanization is a signicant parameter to measure a countrys social and economic progress. It is very important to practice a, reasonable and appropriate, planning and management of cities and towns. The primary step / phase of urban planning is the preparation of present /existing urban land use. In the past conventional methods, the process of preparation of urban landuse is cumbersome involving a lot of labour, material, time and money. Besides, a large scale aerial photographs are being used to extract accurate information. However, aerial photography has limitations and difcult for timely acquisition of data at a desired scale and time. As the urban environment is dynamic in terms of change of land use, expansion, densication and so on, it is essential to have periodic updating of urban land use for review of urban plans for effective implementation. Remote sensing satellites offer excellent opportunities for mapping, monitoring, measuring and managing various urban features. The most attractive aspect of remote sensing is the possibility of repetitive acquisition of imagery over the urban areas for monitoring the changes or developments caused by urbanization and also to help the urban local bodies to enforce regulatory control measures.The launch of IRS series of satellites starting from IRS 1A in March 1988 with coarse resolution (LISS 1 72.5 m, LISS 2 36.25 m) to the present Cartosat-1/ 2 with very high resolution (2.5 m & 1 m) have been providing a scientic information on urban parameters especially the change, growth and land use transformation.

IRS Series Satellites Sensors & Relevance to Urban Planning

There are different levels and types of urban plans,based on the requirement of plan implementation, which can be addressed using the various IRS satellite sensors. Table 1 provides an overview of the satellite data and scale of mapping required for various urban plans/ studies. IRS-1A/1B, the rst of the series of indigenous state-of-art remote sensing satellites carried two cameras, LISS-I and LISS-II with resolutions of 72 m and 36.25 m respectively. It provided a synoptic view of the larger areas on a continuous basis. IRS 1A/1B has been widely used for resource and land use mapping, monitoring urban sprawl, change detection

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Table 1: Satellite data and scale of mapping Sl No. 1 2 3 Type of Map / Planning exercise Regional Plan / Perspective Plan / Urban Sprawl Master Plan / Development Plan / Transport Network Project Plans, Zonal Plans / Large T.P Schemes Scale of Mapping 1:250K 1:50K 1:25K 1:10K 1:5K Data Source (IRS Sensors) 1A / 1B LISS I, LISS II 1C/1D LISS 3, P6 LISS III 1C / 1D PAN P6 LISS IV MX, Cartosat-1 Cartosat-2

and urban demographic studies. Major categories upto Level-1 like built up, transportation, vegetation, agriculture, wastelands, water bodies could be delineated from 1A/1B data. Land use and urban sprawl maps have been prepared for Bombay, Calcutta, Madras, Hyderabad, Ahmedabad and Jaipur (Rao,1991) and in preparation of Regional Plans of Bangalore and National Capital Territory (NCT), Delhi in 1993. IRS-1C/1D is Indias second generation operational remote sensing satellite. These satellites with three cameras (one in Panchromatic band, two in Multispectral band) provided better spatial and spectral resolution (PAN -5.8 m and LISS III - 23.5 m), stereo imaging and improved repetitivity of 5 days and further augmented the remote sensing capability of IRS-1A / 1B. Stereo products have been used in generation of Digital Terrain Models (DTM) which were used for generating of orthorectied products, further enhanced quality of co registration of PAN and LISS III and time series data. The availability of large resolution multispectral data (PAN + LISS III merged) with a wide swath of 70 km, enhanced the capability of extraction of urban features upto Level- II like urban built up, rural settlements, smaller waterbodies and so on. It also allowed to interpret and discriminate a few of Level III classes like industrial areas, recreational, vacant land / layouts. The repetitive capability provided the opportunity to investigate sprawl and change analysis on a scale of 1:12.5K. This has opened avenues to take up the remote sensing data for preparation of base map required for any city / town planning process. It also has been used for circulation network and connectivity planning and location of growth centres. IRS 1C/1D (PAN + LISS III) data have been effectively utilized for the preparation of regional plan 2021 by National Capital Region (NCR), master plan preparation for Hyderabad metropolitan area (2020), route alignment study of power transmission line from Teesta-V-HEP to New Siliguri, Dras - Kargil - Leh, gas pipeline corridor study along Kochi-Mangalore-Bangalore and measuring urban sprawl in Hyderabad (Venugopala Rao et al., 2010). Resourcesat-1 (IRS-P6) launched in 2003, with its multispatial and multispectral sensors has enhanced the capability in carrying out many urban related and infrastructure studies. The multispectral LISS IV MX data with a spatial resolution of 5.8 m provided a good alternative to IRS 1C/1D (PAN + LISS III) merged product. Both the datasets,(LISS IV MX & 1C/1D PAN + LISS III merged) have provided more repetitive coverage, maintained the continuity and provided the temporal & time series datasets, which are useful in detection of urban growth and urban sprawl studies. In land use classication, vacant land class is an important unit, which can be put under proposed land use classes as per the zoning regulations and towards this, the SWIR band of LISS III showed improvements in the delineation of vacant spaces. Detailed land use with eld boundaries at level III and road network up to level II could be extracted from LISS IV MX. (Ramesh et al., 2004). Techniques are being developed for automatic extraction of built up from LISS III datasets of 1C/1D, Resourcesat-1 & 2, which gives the trend of urbanisation of cities /towns. The LISS III data sets have also been used for updation of 1:50K toposheets by Survey of India, Infrastructure studies and landuse change studies by NRSC, Hyderabad. Cartosat-1 & 2 series of satellites has provided us with very high resolution spatial data of 2.5 m and 1 m respectively. It has enabled to generate large scale cartographic and thematic maps on spatial scales of 1:10K and updating capability up to 1:5K scale. Further, it has permitted to observe in detail the urban land use, utility/ infrastructure, growth pattern, open spaces, urban disasters. The 10 bit data offers to identify urban features in their true geometry, shape and

using Cartosat-1 data are (i) Large scale urban land use thematic map and updation on 1:10K scale or better; (ii) Transport network planning; (iii) Urban fringe area monitoring studies; (iv) Green belt/ open spaces mapping; (v) Slum/squatter settlements assessment; (vi) Infrastructure/ utility surveys; (vii) Urban disaster monitoring; (viii) Mapping of heritage/ archeological monuments; (ix) Solid waste/ landll site identication; (x) 3D urban visualization modeling;(Raghavswamy and Pathan,2005).The standard coverages of Cartosat-1 has facilitated operational utilization of satellite imagery like Web services on Bhuvan which enabled the urban local bodies to visualise cities/towns upto Class VI. where the entire possible to generate a baseline DEM product at national level, facilitating drainage network analysis and generation of orthorectied products. These Orthorectied products have become the standard data source in the preparation of base map for city level plans, totally replacing the aerial photographs. It has also helped in generating large scale databases and evolve the land use standards at 10K scale which has changed the scenario and accelerated the urban local bodies to take up preparation of master plans MPs. In this regard, A national scheme, National Urban information System (NUIS) has been initiated by the Ministry of Urban Development. Under NUIS, urban thematic database, at 1:10K scale, have been prepared for 152 Class-I & II towns using Cartosat-1 in conjunction with LISS IV MX. Based on the information collected through the brainstorming session of State Urban Development Departments, Town and Country Planning Departments, Schools of Planning in 2008-10, (http://www. tcpomud.gov.in), the number of approved master plans available in the country is 1233 and about 687 master plans are under various stages of preparation. States like Chattisgarh, Goa, Maharashtra and Gujarat have been extensively using Remote Sensing and GIS Techniques in the preparation of base maps, expediting the process of preparing master plans. The excellent quality stereo images are very useful for 3D terrain mapping and generation of accurate three dimensional maps or 3D city models. Cartosat-2 has not only helped in preparing very large databases (1:5K), for example, Base & Urban landuse of Silvassa, Dadra Nagar Haveli, but also in capturing objects like buildings, structures like water tanks, transmission towers, trafc islands. Different Large Scale Mapping (LSM) projects have been carried out for rural and town planning applications using Cartosat data. 1: 10K scale maps of 74 towns covering 11,500 sq.kms. and 1: 5K scale maps covering 15 class I Municipalities of AP Urban Services for Poor (APUSP), Hyderabad are prepared(http://www.). Cartosat-1 PAN stereo data have been used for East Coast Gas pipeline alignment by generating the land utilization pattern based on primary resource layers and parcel layers. Under Rajiv Awas Yojna (RAY), it is proposed to prepare slum free city plans for over 200 cities in India, wherein Cartosat-1 & 2 data would be used for preparation of GIS database consisting of base features, location and spatial extent of slums. Real estate issues on ownership, encroachments are also being addressed by high resolution Cartosat series data. Remote Sensing data in association with GIS has also provided a non-destructive technique in recognizing historical settlements or archaeological sites. Locating and identifying the ancient site of Dwaraka, settlement sites along the Saraswati river basin using IRS data have been studied in detail (Thakker, 2001). LISS IV MX data has been used to study the ruins of Hampi, Karnataka (Uday et al., 2001). Cartosat-1 and LISS IV MX merged product has been used for archaeological site identication of Nalanda, Bihar and Temple sites like Talakadu, 3D perspective views of Badami (Rajani et al., 2011) country can be visualised with Cartosat. The near instantaneous stereo viewing of Cartosat has made

Major Application Projects

Remote sensing and GIS technologies together provide an efcient and cost-effective means of managing day-to-day activities of town planning, transportation, engineering, facilities 125

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

content on cartographic scales up to 1:25K and on thematic scale better. Major urban applications

management, taxation and good governance. Few major projects carried out by the Remote Sensing Applications Area (RSAA) of NRSC highlighting the capability and usage of various remote sensing satellites are illustrated as follows: Master/Development Plans a. Preparation of Regional Plan-2021 of National Capitol Region in 1999 using IRS1C/1D data and Revision using Resourcesat-1
Fig. 1: Landuse of National Capitol Region in 1993 and 1999

data in 2012. IRS1C (LISS III + PAN) data of 1999

was used to generate the urban sprawl and land use/land cover of National Capital Region (NCR) prepared on 1:50K scale. This was one of the rst GIS database prepared for any metropolitan region and was the primary input for the preparation of Regional Plan-2021. Figure 1 gives the landuse in 1993 and 1999. Presently, for revision of the Regional Plan - 2021, the 1999 GIS database is being updated with Resourcesat LISS IV MX data of 2012 and the land use change statistics have been prepared for Central districts of NCR. These statistics/analysis is being used by Study Groups of Regional Plan revision team for formulating policies and regulations for balanced development of the region. b. Development Plans for Hyderabad i. Hyderabad Urban Development Authority (HUDA) area in 2000 using IRS-1C data: IRS-1C (PAN + LISS-III) merged data (1997 &1998) made possible the creation of rst digital database, for the entire HUDA area, on 1:12.5K scale, consisting of existing landuse and landuse register. Figure 2 shows the landuse from IRS 1D and corresponding master plan. IRS satellite data facilitated updation of physical features and land use within a short time period, with a fairly high degree of accuracy, suitable for land use planning at the metropolitan scale. The database was used towards the master plan preparation of HUDA- 2020. ii. Hyderabad Metropolitan Development Authority (HMDA) Region in 2012 using Carosat-1 and Resourcesat-1 data: HMDA was formed in 2008 by merging the following erstwhile entities: Hyderabad Urban Development Authority (HUDA), Hyderabad Airport Development Authority (HADA), Cyberabad Development Authority (CDA) and Buddha Poornima Project Authority (BPPA) in the year 2008.
Fig. 2: Landuse of HUDA (part) and corresponding master plan - 2020

High resolution Cartosat-1 and LISS IV MX data has been used by HMDA to prepare the

urban geospatial database for the preparation of Development Plan for the metropolitan region (Figure 3). The notied development plan has been published. (http://www.hmda.gov.in ). c. Urban Land use for Dadra & Nagar Haveli (UT) for Development Plan, 2008 using Cartosat-1/2 data:

orthocorrected data of 2007 was used to prepare spatial database for Silvassa town (490 sq.kms.) at 1:10K scale, which was further updated using Cartosat-2 data on 1:5K scale. The Cartosat-2 data enabled to create very large databases suitable for city level planning and was extensively used while creating the Land use register of the town (Figure 4). II. Infrastructure/Facility mapping : a. Land Use/Land Cover Study of Assessment GAILs proposed LNG Pipeline between KochiMangalore-Bangalore, using Remote Sensing and GIS techniques, 2003: IRS-1D (LISS-III) satellite data of 2002 was used to prepare Land Use /Land cover for
Fig. 3: Proposed Landuse zoning of Metropolitan Development Plan-2031 for Hyderabad Metropolitan region

Kochi - Mangalore - Bangalore Gas pipeline, over a distance of 912 km and 1 km corridor width, as shown in Figure 5. The study indicated the total number of crossings (road, rail, canal, river/streams) and predominant land use along the pipeline route and these inputs were used in the Environmental Management Plan (EMP). b. Remote Sensing and GIS based Route Alignment Survey of Transmission Power line between Siliguri (West Bengal) and Teesta (Sikkim) using semi-Automated method, 2004: A feasibility survey of the proposed 400KV D/C transmission line route alignment between Teesta-V HEP in Sikkim to New Siliguri sub-station in West Bengal was carried out using IRS-1D pan sharpened LISS-III geocoded data (2004) and GIS techniques. The satellite data, not only helped to update the map (1972-75 surveyed) details but also provided the recent changes in the settlement patterns, water bodies and vegetation, which was essential while planning the three alternate routes. c. East Coast Gas Pipeline Alignment

Fig. 4: Landuse /Landcover of Silvassa town

using Cartosat-1 Stereo data and GIS Techniques, 2007:

127

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

High resolution Cartosat-1 (PAN)

Cartosat-1 PAN (2.5 m) stereo data (2007), orthorectied with the help of DGPS derived GCPs was used for generating the primary avoidance layers required for nalizing the east coast gas pipeline alignment. The village parcel vectors georeferenced to orthorectified Cartosat-1 PAN data, alongwith the landuse enabled Reliance Gas Transportation Infrastructure Limited (RGTIL) to decide on the cost implications and environmental considerations of the project. Cartosat-1 imagery overlaid with the nal
Fig. 5: Land use / Land cover study

route, village/parcel boundaries and buffers is shown in Figure 6. III. Urban Information Systems a. National Urban Information System (NUIS) Scheme, 2008-2013: For the rst time, at national level, Cartosat-1 data in association with LISS IV MX has been used to generate large scale 1:10K scale multi thematic database for 152 towns. High resolution Cartosat-1 data has enabled the creation of a town level urban database of uniform content and standards across the country. The end utilization of NUIS geospatial data is to support the Urban Local Bodies (ULBs) for GIS based urban planning and governance and develop urban spatial information system to support and strengthen the outreach of geospatial database to SNA and ULB through the use of GIS tools and web services like Bhuvan. In phase-I, out of 152 towns, database creation for 143 towns have been completed and submitted to urban local bodies. Figure 7 gives the NUIS thematic layers at 1:10K scale for Bhavnagar town (Gujarat). In the 12th Five Year Plan, it is being envisaged to prepare the urban thematic database for all the towns up to Class IV using Cartosat-1 & 2 data. IV. Archaeological Studies a. Archaeological Study of Ancient Nalanda Site and Environs, using Cartosat-1 data

Fig. 6: Cartosat-1 image with nal route, buffer and parcel boundaries

was used to locate the present Nalanda archaeological site and determine its extent and precincts. Three striking features were identied on the imagery out of which one mound was nalized for further research. This study established the capability of high resolution Cartosat-1 data in identication, delineation of archaeological sites and also to examine potential sites for further research and excavations Figure 8.

Fig. 7: NUIS thematic layers for one grid (3 x3) of Bhavnagar town

Obtaining requisite information and long term monitoring of urban environment using remote sensing is limited to the available data. Thus, for the last 25 years, remote sensing community has relied on low to high spatial resolution data from the sensors of IRS-1A/1B, IRS1C/1D, Resourcesat-1/2 and Carosat-1/2. The limited spatial resolutions do not allow for detailed thematic information, but provide enough information to analyze regional level urban growth, differentiating urban, non-urban and rural areas. The critical challenges for Indian Remote Sensing Satellites are, very high resolution PAN (0.25 m) and multispectral (1 m) data to allow a higher 2D/3D thematic and higher geometric information for comprehensive urban planning from city to ward level. Urban areas are dominated by impervious

Fig. 8: Nalanda site and three mounds as seen on Cartosat-1

surface material and is an important input for sustainable planning and towards this, high

spatial resolution (1 m) space based Hyperspectral sensors are in need. Urban Heat Island (UHI) is a dominant phenomenon, which has a direct relation to energy demand and human comfort in cities. Mapping and monitoring of UHI by high spatial resolution (1 m) thermal sensors is vital to provide scientic spatial information for preparation of sustainable urban plans. To operationalise the remote sensing technology, high revisit and wide swath remote sensing sensors is a prime requisite. With pixel sizes becoming smaller than the urban features/objects, object oriented classication methodologies, using not only spectral characteristics, but also shape, texture, hierarchical and contextual information is in demand and indispensable. 129

N N R M S

Challenges

B U L L E T I N

M A R C H

2 0 1 3

High resolution Cartosat 1 data

Conclusion
Indian Remote sensing satellites during the last 25 years have decisively conrmed the potential for utilization of various sensors in generation of urban geospatial database at different scales on an operational mode. The important IRS based urban projects such as NUIS, NCR, HUDA, HMDA have adequately established operationalization, for urban applications. State level Urban Development and Town Planning departments have signicantly benetted from IRS datasets in preparing master plans for about 1233 towns out of 5161 census towns (2001),and master plans for 657 towns are under preparation (TCPO, 2011). The continuous development of new sensors in the next generation Indian Remote Sensing satellites will provide more and more advantages in urban areas and support planning.

References
Rao, U.R., (1991). IRS for National Development, Special Issue on Remote Sensing for National Development. Current Science, 61, Nos.3&4, 121-128. Ramesh, B., Rajiv kumar., Neelakantan, K., Shyam Sunder, B., (2004). Land use / Land cover Studies Using IRS-P6 Data. Bulletin of the NNRMS (B) - A special issue on IRS-P6 Data, 29, 67-72. Raghavswamy.V., Pathan. S.K., (2005). Potential of Cartosat-1 High Resolution Data for Urban Applications. Bulletin of the NNRMS (B) - A special issue on Cartosat-1 Data,30,51-57. Rajani, M.B., Rajawat, A.S., Krishna Murthy, M.S., Kamini, J., and Srinivas Rao, (2008). Archaeological exploration in the Talakadu region along the bank of the Cauveri river, Karnataka, Southern India using multi-sensor satellite data and Geographic Information System (GIS). Proceedings of Workshop on Advances in Remote Sensing for Archaeology and Cultural Heritage Management. Rajani, M.B., Patra S.K., and Verma, Mamta, (2011). Space observation for generating 3D perspective views and its implication to the study of the archaeological site of Badami in India. Journal of Cultural Heritage, 10 Rajani M.B., Bhattacharya, S., and Rajawat A.S., (2011). Synergistic Application of Optical and Radar Data for Archaeological Exploration in the Talakadu Region, Karnataka. Journal of the Indian Society of Remote Sensing, 39(4), 519-527. Thakker, P.S., (2001) Remote Sensing of Cosmic cities in Ancient India. Bulletin of the NNRMS - A special issue on Archaeological Applications of Remote Sensing, 26, pp 43-54. TCPO, Outcome report on the Brainstroming Sessions with State Urban Development Departments, Town Plaiing Department and School of Planning, MoUD, Govt. of India, 2011, http://www.tcpomud.gov.in/Divisions/MUTP/ Brainstorming_Report08_10.pdf. Uday Raj., (2001). High Resolution Remote Sensing Data -A Case of Ruins of Hampi. Bulletin of the NNRMS - A special issue on Archaeological Applications of Remote Sensing, 26, pp 61-70. Venugopala Rao, K., Ramesh, B., Bhavani, S.V.L., Kamini, J., (2010) Remote Sensing utility in Urban and Regional Planning, Remote Sensing Applications. NRSC, 109-132. http://www.nrsc.gov.in. http://www.hmda.gov.in/

Introduction
Realizing the large scale benets by having its own space programme to meet national requirements, ISRO developed a dedicated remote sensing satellite series which could provide uninterrupted and assured operational services. Today, India is one of the very few countries to have its indigenous, operational space and ground segment as well as necessary expertise and infrastructure for implementing various remote sensing application projects. Presently, with twelve satellites in operation, IRS is the largest civilian remote sensing satellite constellation in the world, providing images in a variety of spatial, spectral resolutions and swaths to address various types of natural disasters. India is the most ood-affected country in the world after Bangladesh, accounting for one-fth of global death count due to oods (Agarwal and Sunita, 1991). About 40 mha of Indias geographical area is ood-prone (National Flood Control Commission Report, 1980). The chronic ood prone river basins in India are Ganga and Brahmaputra. On account of frequent changes in river courses and braiding of channels, erosion of riverbanks also assumes importance as one of the major problems associated with oods.

IRS-1A Series towards Identifying the Potential

With the launch of Indian Remote Sensing Satellite, IRS-1A, the rst operational remote sensing satellite of the country, ISRO has initiated several studies for showcasing the potential of remote sensing in various needs of ood management in the country (Ramamoorthi et al., 1991). IRS-1A captured the devastating oods in Brahmaputra (in Assam), Ganga (in Bihar), Sutlej, Beas (in Punjab), etc., during 1988. Figure 1 shows the Punjab ood event of 1988 captured by IRS 1A LISS-1 sensor. Further with the launch of IRS-1B and IRS-P2 satellites in early nineties, the potential of remote sensing satellites was demonstrated for mapping the river conguration and embankments of Sutlej rivers, changes in river course of Bhagirathi river in West Bengal, along with ood mapping. The bank erosion maps of Ganga River derived from these satellite datasets were provided to high-level committees on ood management.

IRS-1C Series Towards Operationalisation

The launch of IRS-1C in 1995, has further enhanced the imaging capabilities of ISRO, with WiFS sensor having large swath (810 km) to capture oods in entire State in a single

N N R M S

B U L L E T I N

Srinivasa Rao G, Manjusree P, Bhanumurthy V and Bhatt CM National Remote Sensing Centre Indian Space Research Organization (ISRO), Balanagar, Hyderabad - 500 037, India Email: srinivasarao_g@nrsc.gov.in

M A R C H

SPACE TECHNOLOGY FOR FLOOD MANAGEMENt: 25 YEARS OF IRS

2 0 1 3

shot and frequent repetivity (5 days) for better monitoring of oods, with PAN sensor having high spatial resolution (5.8 m) for better identication of ood control structures and frequent revisit capability (5 days). Together with LISS-3, these three sensors has helped in realising operational services for near real time ood mapping, better identication of ood control structures, etc (Kasturirangan et al., 1996). With the launch of IRS-1D, IRS-P3 in late nineties, the observational capabilities were further enhanced, NRSC / ISRO started several operational services. The major oods and cyclones like Orissa super Cyclone1999, Andhra Pradesh Cyclone-1996, Rajasthan Floods-1996, floods in Assam, Bihar, etc. were mapped and monitored and the information was disseminated to various government departments in near real time mode (Venkatachary et al., 2001).
Fig. 1: IRS-1A LISS-1 image showing 1988 ood inundation in parts of Punjab

Figure 2 shows IRS 1D WiFS image depicting the unprecedented ood situation in Assam as on September, 1998. In addition, data from these sensors were used to study the river bank erosion for Brahmaputra, Ganga rivers and these inputs were used by Central Water Commission (CWC), Ganga Flood Control Commission (GFCC) respectively towards planning flood control measures. Several State and Central Departments realized the use of satellite data for ood management.

Resourcesat, Cartosat, R I S AT S e r i e s To w a r d s Institutionalization

The launch of Resourcesat series and Cartosat series with much better imaging
Fig. 2: IRS 1D WiFS image showing ood situation in Assam as on 11 Sept.1998

capabilities and frequent coverages has paved way for establishment of Decision Support

Centre (DSC) at NRSC under ISRO-DMSP for providing dedicated services for various phases of disasters and particularly the oods. The country started depending on the satellite based inputs from NRSC/ISRO (Navalgund et al., 2007, Roy et al., 2008, Bhatt et al., 2010, Bhatt et al., 2011). Some important events where IRS data was used extensively are Brahmaputra oods (1988 & 1993) in Assam; Orissa Super Cyclone in 1999; Andhra Pradesh Cyclone in 1996, Gujarat cyclone in 1998, monitoring of lake on Pareechu river in Tibet region (2004 / 2005) with potential ood threat to Himachal

to Kosi breach (2008) in Bihar; ash oods in Leh (2010), Jhelum oods (1992, 1993, 1995, 2007, etc) in Jammu & Kashmir; unprecedented Ganga oods (2010) in Uttar Pradesh, Godavari Floods (2006) and Krishna Floods (2009) in Andhra Pradesh; major oods in Assam (2004, 2007, 2012, etc), Bihar (2002, 2004, 2007, 2011, etc), West Bengal (2000, 2007, 2008, etc), Orissa (2001, 2003, 2007, 2008, 2011, etc), Gujarat (2009), Punjab (1988, 2010, etc) and Haryana (2010). The Technology Experiment Satellite (TES) data was used to monitor the potential ood resolution of 1m and high revisit capability of 1 day. The information was provided to Ministry of Home Affairs (MHA), Cabinet Secretarys Ofce on daily basis initially and at regular intervals subsequently till its breach in 2005. The launch of RISAT series, with new capability of imaging during day & night and during adverse cloud conditions, has further increased the ood and cyclone monitoring capabilities. The major oods in Assam during 2012 and the inundation due Cyclone NILAM in Andhra Pradesh are the recent examples. NRSC / ISRO has been providing, free of cost, operational ood information services to several States through State Remote Sensing Application Centres (SRSACs), Flood Management Improvement Support (FMIS) Centre, etc. In addition, capacity building is also imparted to several State Agencies.

Flood Mapping and Flood Inundation Products

During initial phases, visual interpretation and manual map preparation techniques were used to show case the potential of satellite remote sensing for ood management studies, resulting in a turn-around -time of 72 hours to generate ood inundation maps. Subsequently, to improve turn-around-time, digital techniques were introduced for satellite data analysis, extraction of ood inundation layer and generation of ood products and computation of ood inundation statistics. With these techniques, the turn-around-time has become about 6 hours in most of the cases. Flood maps are composed at state / district level and for selected areas at detailed level (Figure 3). The state-level ood map shows ood inundation layer, overlaid with state and district boundaries, major roads and railway, whereas in the district ood map submerged road/rail and taluk boundaries are also shown. For selected areas in a district, detailed ood maps at 1:50k scale showing flood inundated areas, submerged road/ rail overlaid with village boundaries are also composed. A brief ood report containing the district-wise ood inundated area statistics along with different ood inundation maps is disseminated to the state agencies. In addition, towards better outreach, flood layer in GIS format is also provided to State Remote Sensing Application Centers, FMISCs for further value addition and providing
Fig. 3: Flood inundation products

information to line departments.

133

N N R M S

B U L L E T I N

threat to Himachal Pradesh in 2004 due to the lake on Pareechu river in Tibet region with high spatial

M A R C H

2 0 1 3

Pradesh; the devastating Tsunami (2004) over Andaman islands, Tamilnadu and other areas; oods due

Spatial inundation maps, along with estimates on submergence are disseminated to the Ministry of Home Affairs, Central Water Commission and State Relief Commissioners through electronic means for further use in relief and rescue operations. Flood maps depicting the ancillary information like administrative boundaries, settlements, roads and rails, are being widely used for carrying out relief and rescue operations specically for airdropping relief material.

Appreciation from Users

Several organizations appreciated the unrelenting efforts of NRSC / ISRO in supporting the disaster management activities of respective States. Appreciation received from Joint Secretary Dept. of Agriculture and Co-operation, for the support extended during 1988 Assam Floods, Honble Chief Minister of Orissa for the support during Orissa Super Cyclone in 1999, Joint Secretary, Ministry of Home Affairs for the support during Pareechu and Assam Floods in 2004, Govt of Bihar for the support during Kosi Floods in 2008, Govt of Assam for the support during Assam Floods 2012, are to mention a few.

Recognition of IRS Services at National Level

These unrelenting efforts in ood management has given NRSC/ISRO a unique identity at National level. Today with a constellation of satellites operating in variety of spatial resolutions, spectral bands and swaths India is in a position, not only to respond to major ood disaster event in the country, but playing key role in providing space based inputs for ood mitigation like ood hazard zonation. For the rst time in the country, NRSC/ISRO has prepared Flood Hazard Atlas of Assam State based on 10 years of satellite based observations. This was released by Honble Chief Minister of Assam in 2011 (Sharma et al., 2012). Preparation of similar ood hazard atlas for Bihar, Orissa, West Bengal, etc., are under progress. Satellite images have become an integral part for different phases of ood disaster management cycle i.e. pre ood (preparedness), during ood (relief and rescue operations) and post ood (mitigation measures) in the country. National Disaster Management Authority (NDMA) the apex body for the management of disasters in the country, has issued guidelines for ood management which includes the utilization of remote sensing data in different aspects of ood management.

IRS Support for Other Countries

Apart from fullling the national needs, IRS series of satellite have captured several major disasters in other countries. The oods in Pakistan (2010), Thailand (2011), Philippines (2012), Cyclone Bopha in Philippines (2012), Japan Tsunami (2011) are some of the recent examples. This gave an unique opportunity to serve other countries through various international forums like International Charter, Sentinel Asia, UNOOSA, UNSPIDER and UNESCAP. Following are some of the success stories of IRS applications to Flood Management in the country. Assam Floods, 2012 Assam witnessed one of the devastating floods in last week of June, 2012. Data from Resourcesat, RISAT series was used extensively along with other satellite datasets to help the Assam State Disaster
Fig. 4a: Resourcesat-2 image showing ood inundation around Kaziranga National Park area as on July 01, 2012

Management Department for planning the relief operations. The information includes

submerged roads, rails, villages, etc., in the form of maps and statistics. This information was also used by the State Govt. to apprise the Honble Prime Minister of India about the ood situation. Figure 4a & 4b show the
Fig. 4b: RISAT-1 & 2 images showing ood situation and affected railway track in parts of Dibrugarh district

ood situation in Kaziranga and Dibrugarh areas in Assam

Leh, Jammu & Kashmir Flash Floods, 2010

Flash ood and mud slides were triggered due to a cloudburst that occurred near Leh in Jammu and Kashmir on August 6, 2010 causing extensive damage. High resolution IRS data from Cartosat-1, Cartosat-2 and Cartosat-2B was used for identication of the damage to roads, houses, canals and agricultural elds, etc., especially around Choglamsar village. Figure 5 shows mudow and damges near Leh town.

Bihar Floods, 2008

Fig. 5: Mudow and damges near Leh town

Kosi, a trans-boundary river, ows into Bihar, India from Nepal. A sudden breach in Nepal devastated several districts of east Bihar during August 18, 2008. As this was an unusual event, the State Govt needed more information from NRSC / ISRO. Again, IRS series of satellites provided information on the ood progression and recession at regular intervals, which helped the State Government for managing the situation in a better manner. Information provided almost on alternate days from IRS and other satellites was used by several departments for aerial inspection of ood affected area, identification of inundated villages and planning for distribution of relief materials. Govt of Bihar appreciated the support extended by NRSC during the Kosi oods. Figure 6a & 6b show the pre and post ood inundation caused due to the breach and the location of the breach.

Fig. 6a: IRS P6 AWiFS satellite image showing pre breach (11-April-2008) and post breach (20-Oct-2008) situation

135

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

extent of ood inundation in various districts,

Floods Due to Lake Burst: Pareechu Lake-2004

Floods were reported due to burst of lakes on trans boundary rivers, causing extensive damage in India. The ash oods due to lake outburst on Brahmaputra River in Tibet caused severe damage in Arunachal Pradesh during 2000. Yet in another incident, the lake outburst on Sutlej River in Tibet caused extensive damage to Himachal Pradesh, particularly to Naptha Jhakri Power Project. Again, similar lake formed in Sutlej basin, Tibet in 2004. As the lake is formed
Fig. 6b: High resolution Cartosat-2 image (08-Sept-2008) showing the breach location

in a inaccessible area satellite data is the only source for identifying and monitoring the lake. NRSC/ISRO has used IRS data extensively for this purpose. NRSC has helped MHA and Govt. of Himachal Pradesh by monitoring the lake status on day to day basis. This has emphasized the utility of satellite technology in monitoring of lakes. Figure 7 shows the building up of Pareechu lake during Jun-July, 2004.

Tsunami 2004
IRS data was extensively used during
Fig. 7: IRS images showing building up of Pareechu lake during Jun-July, 2004.

The Great Indian Ocean Tsunami, which hit Indian coast on December 26, 2004. IRS captured the Tsunami devastation over parts of Andaman & Nicobar islands, just few hours after the tsunami. Again, NRSC / ISRO has tasked all IRS satellites and quickly provided the impact of the tsunami over A&N and other States. Figure 8 shows effect of tsunami over Trinkat Island.

Super Cyclone Orissa, 1999

A super cyclonic storm hit the Orissa coast on October 29, 1999 causing
Fig. 8: Comparison of AWiFS images of Pre-Tsunami and post-Tsunami event over Trinkat Island

devastating damage to several coastal districts in Orissa State. IRS data was acquired to provide assessment of affected areas, damage

to crops and properties to the state agencies. Inundation maps were delivered physically as there was no internet available during those days. This event has further increased ISROs resolve towards disaster management support in the country. Figure 9 shows the impact of Orissa Super Cyclone over coastal districts of Orissa through a series of multi-temporal IRS satellite images.

administrators and planners to identify areas that are frequently ooded and prioritize their mitigation efforts. Using IRS and other satellite datasets acquired during various ood events from 1998-2007, ood hazard atlas was generated based on frequency of ooding. This is rst of its kind and will be useful to planners. to identify inundated areas with different frequencies and better ood
Fig. 9: Multi-temporal IRS satellite images showing the inundation recession post Orissa super cyclone.

ood hazard map for Assam state.

River Bank Erosion and Deposition: Ganga River near Chappra in Bhojpur, Bihar
IRS potential for river bank erosion studies were showcased for Ganga, Brahmaputra and other rivers using IRS satellite datasets. These inputs were provided to high level committees on flood management for deciding on anti erosion measures. Figure 11 shows the bank erosion and deposition identied near Chappra in Bhojpur district, Bihar along River Ganga using IRS-1A LISS-II data of 1990 and IRS-P6
Fig. 10: Flood hazard map of Assam

LISS-III data of 2004.

Conclusions
IRS satellites are the mainstay for the Disaster management activities especially ood monitoring. The ood monitoring activities initiated during the IRS 1A period has matured through the period and currently, all the major oods in the country are being monitored and the near real time information is being provided to the concerned for immediate actions. In addition to the monitoring, hazard zonation based on the historical database, and river bank erosion monitoring are also being carried out. With the data from the IRS series of satellites, in addition to the ood events in the country, global disasters are also being supported.

Acknowledgement
The authors gratefully acknowledge the support and guidance provided by the Director, National Remote Sensing Centre, Hyderabad and the Deputy Director, Remote Sensing Applications Area, National Remote Sensing Centre, Hyderabad. The authors would also like to thank all the scientists involved in the analysis of satellite images and generation of various types of outputs for the case studies discussed in the above sections. 137

N N R M S

management in the state. Figure 10 shows

B U L L E T I N

M A R C H

Flood hazard maps facilitate the

2 0 1 3

F l o o d H a z a rd Z o n a t i o n Assam

References
Agarwal, A., Sunita N, (1991). Book on Floods, Flood Plains and Environmental Myths. Bhatt, C.M., Rao, G.S., Manjusree, P. and Bhanumurthy, V. (2010). Space Based Disaster Management of 2008 Kosi Floods, North Bihar, India. Journal of the Indian Society of Remote Sensing, 38, pp. 99108. Bhatt C.M., Rao G.S., Manjushree P. and Bhanumurthy, V. (2011). Potential of High Resolution Satellite Data For Disaster Management: A Case Study of Leh, Jammu & Kashmir (India) Flash Floods, 2010, Journal of Geomatics Natural Hazards and Risks, Vol.2(4):365-375.
Fig. 11: River Erosion and Deposition in Ganga River near Chappra in Bhojpur, Bihar

Kasturirangan, K. Aravamudan,R., Deekshatulu, B.L., Joseph,G. and Chandrasekhar, M.G. (1996). Indian Remote

Sensing satellite (IRS-1C)-The Beginning of A New Era. Current Science 70 (7), 495-500. National Flood Commission Report (Rashtriya Barh Ayog), 1980 Navalgund, R.R., Jayaraman, V. and Roy, P.S. (2007). Remote Sensing Applications: An Overview. Current Science, 12 (25), 1747-1766. Ramamoorthi, A.S., Thiruvengadachari, S. and Kulkarni, A.V. (1991). IRS-1A Applications in Hydrology and Water Resources. Current Science, 61 (3&4), 180-188. Roy, P.S., Bhanumurthy, V., Murthy C.S. and Kiran Chand, T. (2008). Space For Disaster Management: Lessons and Perspectives. Journal of South Asian Disaster Studies (1), 157-177. Sharma SVSP, Srinivasa Rao G, Bhatt CM, Manjusree P & Bhanumurthy V (2012). Development of ood hazard maps for Assam state, India using historical multi-temporal satellite images. NNRMS (B)-36, JUNE 2012, PP87-92 2. Venkatachary, K.V., Bandhypadhyay, K., Bhanumurthy, V., Rao, G.S., Sudhakar, S., Pal, D.K., Das, R.K., Sarma, U., Manikaran, B., Rani, H.C.M. and Srivastava, S.K. (2001). Dening a space-based disaster management system for oods: A case study for damage assessment due to 1998 Brahmaputra oods. Current Science. 80 3, 369-377.

NAtIONAL AGRICULtURAL DROUGHt ASSESSMENt & MONItORING SYStEM (NADAMS): GENESIS & WAY FORWARD
Sesha Sai MVR, Murthy CS, Chandrasekar K and Raghavaswamy V National Remote Sensing Centre (ISRO), Hyderabad - 500 037, India Email: seshasai_mvr@nrsc.gov.in

Introduction
Drought is a complex natural phenomenon and its assessment is dependent on a number of interlinked parameters. Because of the complexities of drought, no single index has been able to adequately capture the intensity and severity of drought and its potential impacts. Different states are adopting different methods and criteria for drought assessment, drought declaration, preparation of drought memorandums and relief assessment etc. The three types of drought viz., meteorological, hydrological and agricultural drought are interconnected to some extent. Meteorological drought is the root cause of drought leading to hydrological and agricultural drought. Immediate impact of drought is perceived on agriculture sector. Agricultural drought occurs when the soil moisture availability is inadequate to support normal crop growth. Reduction in crop sown area, delay in the time of crop sowing, stunted crop growth and reduced crop yield are the manifestations of agricultural drought. Based on time of occurrence, agricultural drought is classied as early season drought, mid-season drought and terminal drought. Agricultural drought results from the complex interactions between weather, soil, crop and human actions and hence, the assessment of the intensity of agricultural drought continues to be a challenging task for researchers, drought managers and policy makers. Meteorological drought is measured by rainfall data recorded by weather stations and hydrological drought is assessed by inows into the surface water bodies measured through gauging points. But the agricultural drought assessment could not be accomplished by such direct and quantitative measurements. It requires quantitative information related to soil moisture, planting pattern and crop condition along with their inter-relations which is not adequately provided through manual surveys or sparsely located measuring stations. Satellite remote sensing has addressed this issue to some extent through crop stress detection and crop condition monitoring in near real-time through bio-physical indicators derived from satellite data. For more than two decades, coarse resolution satellite sensors are being used routinely to monitor vegetation and detect the impact of moisture stress on vegetation (Thenkabail et al., 2004, Minamiguchi Nokio, 2005).

Evolution of NADAMS
After the phenomenal drought during 1987 which affected most parts of our country, the then National Remote Sensing Agency (NRSA) started research activities to monitor agricultural drought using space technology. The sustained research efforts evolved

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

a national level drought monitoring system named as National Agricultural Drought Assessment and Monitoring System (NADAMS). NADAMS initiated by NRSA was initially funded by the Department of Agriculture and Cooperation with support from State and Central Government departments. Subsequently, NADAMS was fully sponsored by the Department of Space. From 2004, NADAMS was brought under Disaster Management Support Program (DMSP) of ISRO. During the period 1987 to 2012, NADAMS has undergone many methodological improvements, keeping pace with space technology developments and user requirements. During the initial years, daily data of NOAA AVHRR HRPT data (1.1 km) was analyzed to generate the biweekly vegetation index and provided periodic information on crop condition at district level. NOAA AVHRR has a unique advantage of large swath (c.a. 2800 km) and high temporal coverage. The NDVI derived from NOAA AVHRR data was extensively used for vegetation monitoring and was reported to be a potential dataset for agriculture monitoring (Malingreau 1986, Kogan 2001). Therefore, AVHRR NDVI was put in to use in NADAMS as a satellite based drought indicator. The operational services from NADAMS commenced during kharif 1989, with the issue of biweekly interval drought bulletins from June to November. The bulletins provided colour coded AVHRR NDVI images of current season and previous normal season, vegetation condition map showing agricultural drought condition in terms of normal, mild, moderate and severe during the respective fortnight. The bulletins were issued to all the concerned district and state departments with effective turnaround time of 2-3 days for every biweekly period. Subsequently from 1992 onwards instead of biweekly bulletin, monthly detailed reports on crop and seasonal conditions were issued at district level. This report provided compilation of rainfall and satellite based crop condition assessment at district level for thirteen agriculturally important and drought vulnerable states of the country. Thus, during the period 1989-1995, NOAA AVHRR 1 km data was the main satellite dataset used in NADAMS and the scale of assessment was district level.

Utilization of IRS Data in NADAMS

Launch of the second generation satellite viz., IRS-1C during December 1995 with a new sensor Wide Field Sensor (WiFS) having a spatial resolution of 188 m, temporal repetivity of 5 days, optimum spectral and radiometric response had brought the most noteworthy change in the drought assessment perspective of NADAMS. WiFS data utilisation had culminated into two unique additions to NADAMS improved scale of assessment from district to subdistrict level and more focused assessment from total geographic area monitoring to agricultural area monitoring. After experimental endeavor in kharif 1996, operational assessment of taluk level agricultural drought was commenced in kharif 1997 in two states namely Andhra Pradesh and Karnataka. With the launch of IRS 1D and subsequent availability of its WiFS data from 1997, the sub-district level agricultural areas monitoring was further strengthened. Combined use of IRS1C WiFS and IRS1D WiFS had improved the temporal revisit to about 3 days. Thus, WiFS data utilization was successfully adopted in NADAMS during the period 1996-2003 leading to signicant improvements in the drought assessment and operational services to the user community. A major boost to the detailed monitoring of agricultural drought had occured after the launch of IRS-P6/ Resourcesat 1 satellite during October, 2003. It carried Advanced Wide Field Sensor (AWiFS) which has a spatial resolution of 56 m with four spectral bands including a SWIR band (1.55 to 1.70 m). With the utilization of AWiFS sensor data, the sub-district level drought assessment had improved in a remarkable way (Murthy et al., 2007). Improved spatial resolution resulted in more focused monitoring. Improved spectral resolution with added SWIR band enabled generation of one more index Normalized Difference Water Index (NDWI) for drought assessment. The availability of SWIR band in Resourcesat-1 enabled the derivation of a crop water content sensitive index called NDWI, which is very useful for monitoring the water stress during the agricultural season. Use of NDVI and NDWI have strengthened the drought assessment. Considering the potential of WiFS and AWiFS datasets, the sub-district level assessment was extended to two more states namely Maharashtra and Haryana in 2004. Thus, by 2004, four states were covered with sub-district level agricultural drought

of sub-district level agricultural drought assessment in the above four states.

Combined use of AWiFS and WiFS data, practiced in NADAMS from 2004-2006, permitted increased frequency of images and reduced the cloud contamination in the time composite NDVI images. Inter sensor normalization studies were carried-out to normalize the differences in the sensors data. Normalization was done between (1) IRS1C WiFS and IRS1D WiFS, (2) IRS1D WiFS and Resourcesat 1 with multiple pairs of concurrent data of red and NIR bands as well as NDVI. These regression functions were used to normalize the NDVI differences between two sensors. Studies were also carried out on the comparison of WiFS/AWiFS NDVI with that of MODIS and found the compatibility between the two datasets. AWiFS and (3) Resourcesat 1 AWiFS and Resourcesat 2 AWiFS. Regression functions were developed

Evolution of Methodology in NADAMS

The methodology of NADAMS had progressively improved from time to time keeping pace with the research and development results. Till 2003, NDVI was largely used for drought assessment. NDVI is a robust and proven indicator to bring-out the changes in crop condition and vigour. After 2003, the methodology had been continuously augmented with new indices and bio-physical products. With the launch of Resourcesat-1 in 2003, SWIR based NDWI had complemented the existing NDVI based assessment. MODIS data utilization had started in 2004 and enabled generation of both NDVI and NDWI at 3 spectral resolutions 250 m, 500 m and 1000 m. Normalised Difference Water Index (NDWI) is derived from two bands including a moisture sensitive SWIR band and insensitive NIR band. In the beginning of the cropping season, when soil background is dominant, SWIR is sensitive to surface wetness of top soil. As the crop progresses, SWIR becomes sensitive to leaf moisture content. When the crop is grown up, the response in SWIR band is mostly from canopy and not from the underlying soil. NDWI has been a popular index for crop stress detection and for monitoring moisture condition of crop/vegetation canopies over larger areas (Fensholt and Sandholt 2003). The response of NDWI to moisture is instantaneous without any time lag. NDWI is more sensitive to both desiccation and wilting. NDVI is more sensitive to canopy chlorophyll changes and tend to saturate at high biomass levels. In view of the limitations associated with individual indicators either NDVI or NDWI, combination of both the indicators may provide a robust approach for drought monitoring. Studies carriedout in NADAMS indicated complementary relationship between NDVI and NDWI anomalies and hence the combination of NDVI and NDWI anomaly classes would lead to generate additional information on crop condition and crop stress. The decision rules based on NDVI and NDWI anomalies were developed and are being implemented for drought assessment. Images of NDVI and NDWI of rst fortnight of October 2012 in Figure 1 show that in parts of Andhra Pradesh, Karnataka and Rajasthan appear green in NDVI while it appears yellowish in NDWI. This can be attributed to the sensitivity of NDVI to the standing crop green canopy while NDWI is sensitive to the moisture stress the crop underwent due to lack of rainfall. Thus NDWI is very useful in identifying the crop liquid water stress particularly in the early and late part of the season. In addition to NDVI and NDWI, geospatial products like AMSR-E soil moisture, TRMM Rainfall, NOAA CPC rainfall forecasts are being used for drought assessment in NADAMS. From 2010, an angular

141

N N R M S

B U L L E T I N

M A R C H

Inter Sensor Normalization Studies

2 0 1 3

assessment. Availability of Resourcesat-2 AWiFS data from August 2011 resulted in the continuation

index Shortwave Angle Slope Index (SASI) derived from NIR and two SWIR bands has been adopted in NADAMS. SASI is more sensitive to surface wetness/dryness and hence is a potential index for detecting early season drought conditions. The emphasis on integrated approach towards developing a robust assessment procedure in NADAMS had started in 2006 and by 2008 an operational approach with a combination of multiple
Fig. 1: NDVI and NDWI image of the country during October 2012

indices, bio-physical products, augmented ground data base on rainfall and crop sown area was put into practice

Angular Indices
An advantage of angle indices is that they are relatively insensitive to albedo differences for comparison between spectra. An angular new index - Shortwave Angle Slope Index (SASI), adopted from Khanna et al., (2007), is being used in NADAMS from kharif 2010. SASI is calculated as product of the angle at SWIR1 1240 nm (SWIR1) and the inclination of the line that passes through the NIR and SWIR 2 at 1640 nm of a triangle with vertices at R(NIR), R(SWIR1) and R(SWIR2), where R is the reectance at broad bands. The slope of line, (RSWIR2RNIR) / (SWIR2NIR) can be approximated by the difference of the reectance at NIR and SWIR2 since the wavelength difference between the two vertices is constant. MODIS 500 m 8-day composite images of NIR, SWIR 1 and SWIR 2 spectral bands are being used for computing the index. SASI is very sensitive to surface wetness. Unlike other vegetation indices SASI considers the reectance values at certain wavelength as well as their interaction in terms of angle and slope. Because of the involvement of SWIR bands, the index is less affected by the path radiance. It also transforms its sign from positive to negative, once the major portion of the surface background changes from soil to vegetation. Prabir et al., (2011) demonstrated that weekly SASI profiles at district level revealed seasonality and strong association with rainfall and sown area pattern. A typical SASI image over India is shown in Figure 2a. Chronological synchronization was observed between (a) time of rainfall increase, (b) time of SASI decrease and (c) time of crop sown area increase in the monsoon season (Figure 2b). Positive to negative transformation of SASI in the seasonal prole provide unique interpretation of dryness to wetness shift and this property of SASI can be exploited to assess the area favourable for crop sowing. In the absence of operational procedures for nearrealtime soil moisture estimation, SASI can act as surrogate parameter to draw inferences on
Fig. 2a: SASI image over Indian region during July 2012

the commencement and progression of crop

situation in the early part of the season. Using multiple criteria SASI, soil texture, rice/non-rice areas, a soil water balance based procedure was evolved to generate spatial product on crop sowing favourable area or already cropped area in the for Crop Sowing / Crop Sown Area (AFCS), has been generated on fortnightly or monthly scale from June-September. After validation with state level crop sown area statistics reported by respective agriculture
Fig. 2b: SASI and its relationship with rainfall and crop sown area during kharif season

departments, the product is generated and used operationally from kharif 2010. The cloud covered pixels in SASI images are resolved with the support of rainfall and water balance derived soil moisture index. Extending SASI for rabi season monitoring, (Murthy et al., 2012), mapped spatial patterns of surface wetness and rice transplantation in rabi season of recent 10 years (2002-03 to 2011-12) in a rice dominant irrigated command area.

Fig. 3: AMSR-E surface soil moisture of the country during June-July 2007

Soil Moisture Monitoring Using Passive Microwave Data

Soil moisture plays a key role in the success of the crop. Hence, monitoring of soil moisture is very important for the success of agricultural drought monitoring. In order to capture the surface soil moisture status, passive microwave AMSR-E sensor derived soil moisture products were used in the project (Chakraborty et al., 2012). Figure

Fig. 4: Top 30 cm soil moisture derived from SWB model during kharif 2012

3 shows the AMSR-E surface soil moisture during July 2007 which clearly bring out

the region which has saturated moisture content and decit moisture content. Passive microwave soil moisture is very useful in early part of the season when spectral based indices has limitations in providing information.

Development of Soil Water Balance(SWB) Model

Passive microware data provides surface soil moisture upto a depth of 5 cm which is adequate during early part of the season. To effectively monitor crop throughout the season, knowledge on top 30 cm and root zone soil moisture is very important to assess the performance of the crop (Figure 4). 143

N N R M S

B U L L E T I N

season. This product called Area Favourable

M A R C H

2 0 1 3

sowings and to characterize the agricultural

A soil water balance model was developed under NADAMS which was run on a daily basis to provide the top 30 cm soil moisture and root zone soil moisture for some selected crop throughout the kharif season. The top soil moisture is very useful input in deriving the area favorable for crop sowing during early part of the season. The root zone soil moisture for the selected crops is helpful in identifying the exact period and duration of stress in a particular region. The soil water balance model has been a very important input in the assessment and monitoring of agricultural drought during the season.

Present Status
Agricultural drought monitoring under NADAMS has been carried out by NRSC for more than 2 decades. Adoption of proven indices and sound methodology has made NADAMS assessment more objective and user oriented. The reports of NADAMS form a very important input to the decision makers at the Ministry of Agriculture, State Agriculture and Relief Departments. Realizing the importance of drought information and maturity and soundness of NADAMS project, Department of Agriculture and Co-operation (DAC) under Ministry of Agriculture decided to execute the operational activities of the project through its recently inaugurated Centre - Mahalanobis National Crop Forecast Centre at Delhi with the technical support of NRSC, ISRO. NRSC undertook methodology transfer, database transfer and training to MNCFC. All the activities of NADAMS in kharif 2012 were carried out by MNCFC. Drought impact assessment, early warning, exploration of new indices, development of new bio-physical products are some of the issues to be addressed to take the drought monitoring endeavor to next level and to achieve resilient rainfed agriculture in India.

Conclusion
NADAMS of NRSC successfully incorporated different satellite derived indices, geospatial products to represent soil moisture and crop status during the season, showcased the operational robustness of geospatial approach and strengthen the drought assessment mechanism in the country. NADAMS had undergone many methodological improvements since its inception in 1987. WiFS and AWiFS data from IRS satellites have brought a remarkable change in the drought assessment perspective of NADAMS. The indices and their anomalies derived from WiFS and AWiFS are found be compatible with that derived from other sensor data like MODIS, thus enabling the synergistic use of IRS data with other available global datasets. The transfer of NADAMS to user department MNCFC of Ministry of Agriculture, Government of India is a success story of yet another remote sensing application project. Drought impact assessment and early warning are some of the yet-to-be operationalised issues to take the drought monitoring endeavor to next level and to achieve resilient rainfed agriculture in India.

Acknowledgements
We express our sincere gratitude to Dr. V.K. Dadhwal, Director, National Remote Sensing Centre for his constant encouragement and guidance. Valuable suggestions and guidance offered by Dr. V. Raghavaswamy, Deputy Director, Remote Sensing Applications, NRSC are sincerely acknowledged.

References
Chakraborty, Abhishek, Sesha Sai, M.V.R., Murthy, C.S., Roy, P.S., Behera, G., (2012). Assessment of area favourable for crop sowing using AMSR-E derived soil moisture index. International Journal of Applied Earth Observations and Geoinformations, 18.pp 537-547 Fensholt, R., and Sandholt, I., (2003). Deviation of a shortwave infrared water stress index from MODIS near and short wave infrared data in a semi-arid environment, Remote Sensing of Environment, 87, pp. 111-121.

Developement of angle indexs for soil moisture estimation, dry matter detection and land-cover discrimination. Remote Sens. Environ. 109, pp. 154-165. Kogan, F.N., (2001). Operational Space Technology for Global Vegetation Assessment, Bulletin of the American Meteorological Society, 82, pp1949-1964. Malingreau, J.P., (1986). Global vegetation dynamics: Satellite observations over Asia, International Journal of Remote Sensing, 7, 1121-1146. Minamiguchi Nokio, (2005). The application of Geospatial and Disaster information for food insecurity and agricultural drought monitoring and assessment by the FAO GIEWS and Asia FIVIMS, Workshop on Reducing Food Insecurity Associated with natural Disasters in Asis and the Pacic, Bangkok, Thailand, 27-28 January 2005. Murthy, C.S., SeshaSai. M.V.R., BhanujaKumari. V., and Roy, P.S. (2007) Agricultural drought assessment at disaggregated level using AWiFS/WiFS data of Indian Remote Sensing satellites, Geocarto International, 22, pp127-140 Murthy, C.S., Abid, S.M, Prabir Kumar Das, Sesha Sai, M. V. R, (2012). Tracking surface wetness and rice transplantation using Shortwave Angle Slope Index (SASI), Proc. of ISRS Symposium, New Delhi 2012. Prabir Kumar Das, Murthy, C.S, Sesha Sai, M.V.R., and Behera, G. (2011). Analyzing seasonal dryness/ wetness conditions during crop sowing period using Shortwave Angle Slope Index. Proc. ISRS Symposium, Bhopal, 2011 Thenkabail, P. S., Gamage, M.S.D.N., and Smakhtin, V.U., (2004) The use of remote sensing data for drought assessment and monitoring in Southwest Asia, International Water Management Institute (IWMI) Research Report no.85, Srilanka.

145

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

Khanna, S., Palacios-Orueta, A., Whiting, M. L., Ustin, S. L., Riano, D., and Litago, J., (2007).

CAPACItY BUILDING IN EARtH ObSERvAtION TECHNOLOGY AND APPLICAtIONS: ACHIEvEMENtS AND FUtURE CHALLENGES
Team IIRS Indian Institute of Remote Sensing Indian Space Research Organisation, Dehradun 248 001, India Email: director@iirs.gov.in

Introduction
Indian Space Research Organisation (ISRO) under the Department of Space, Govt. of India, has created a world-class space infrastructure primarily for socio-economic development of the human kind in the country. With a modest beginning in mid-seventies by designing and launching a few experimental satellites, India today has a large constellation of satellites dedicated towards Earth Observation(EO), communication and broadcasting, meteorology, navigation and space science missions. The launch of IRS-1A (Indian Remote Sensing Satellite-1A) on 17th March 1988 set the era of operational EO programme not only for natural resources survey and management but also for infrastructure development planning, disaster risk management, environmental applications, etc. With the completion of two and a half decades of several successful IRS missions, a range of EO data in terms of spectral bands, spatial and temporal resolutions, swath and stereo-capability have now become available and used for generating geo-information (GI) at different scales for characterizing spatial processes and developing applications for socio-economic benets. The National Natural Resources Management System (NNRMS), a national level inter-agency system set-up in 1983 and chaired by the Member, Planning Commission, Govt. of India, steers the EO programme in the country. The nine Standing Committees on Agriculture & Soils, Bio-resources, Cartography & Mapping, Geology & Mineral Resources, Ocean & Meteorology, Rural Development, Training & Technology, Urban Development, Water Resources chaired by the Secretaries of the concerned Ministries of the Govt. of India, provide vital inputs on the requirements of space, ground and user segments. Over the last two decades, EO data from IRS missions have proved vital in generating the geospatial information on various natural resources, viz. land use/ land cover, surface and ground-water, snow/ ice cover, wasteland, biodiversity, potential zones for shing and many more. The EO data obtained from Indian satellites and the thematic information generated through them are shared with public through a web-portal, called BHUVAN (www.bhuvan.nrsc. gov.in), for visualization and geospatial analysis for developmental planning. Capacity Building (CB) of human resource is recognized as a critical element to strengthen not only the organization/ institutions but even the people at grassroots level by spreading awareness about the power of GI in spatial planning. There is a large gap between demand and availability of trained human resource, who can benecially utilize the geospatial information. Such trained and educated human resource is needed both as the provider and user of GI. This is vital for sustainable development so that informed decision making can be done at all levels. Further, with the advancements of technology and newer systems and data coming through current and future

addresses this particular issue and provides overall direction and guidance for capacity development in the country, so that, the benets of the space technology reach the society at large. The twelfth Five Year Plan (FYP) document (http://planningcommission.gov.in/), released on December 21, 2012, also emphasizes that development of human and institutional capabilities and infrastructure are of paramount importance for faster, more inclusive and sustainable growth of the country. With the completion of 25 years of EO through IRS missions, this article attempts to give particularly in the light of future Indian EO missions and the Indian Governments focus on capacity building initiatives under the twelfth FYP.

Capacity Building Global and National Scenario

The study of EO data is a need of todays world to keep account of the existing natural resources and also to maintain the environment for better sustainability. There is a growing need for professionals who can use the EO data along with allied geo-ICT to prepare, maintain and update geographic databases and develop new geoinformatics tools. There has also been an increased awareness in government and non-government sectors on generation and use of such geo-databases. Therefore, capacity building of the human resources at all levels has been recognised as the key area needing immediate attention. Global Scenario The capacity building in the eld of EO technology and its applications is carried out through training and education programmes world over. The programmes are offered at postgraduate and undergraduate level either as specialized courses or as one of the subjects in various courses of science or engineering discipline. Majority of the courses are offered at postgraduate level; a few countries (e.g. Australia, Canada, China, Denmark, Germany, India, Sweden, Switzerland, USA) also offer courses at undergraduate level. It is important to note that the EO technology and applications being the multidisciplinary eld, the courses/ programmes are being offered as different names, viz. Remote Sensing (RS), RS & GIS (geographical information system), geoinformatics, geomatics, geoinformation science and EO, geospatial/ spatial technology, etc. According to Rhind and Raper (2004), globally about 2000 universities offer the courses in this eld at undergraduate and postgraduate level. Additionally, a huge number of non-academic organizations across the world, particularly software companies, also offer such courses.There are more than 100 institutions/ universities/ software companies in USA alone offering undergraduate and postgraduate courses in this eld (http://spatialnews.geocommunity. com/ education/links/usa.html). Amongst the Asian countries, China tops the list with
Fig. 1: IIRS main building

over 100 four-years engineering degree

147

N N R M S

B U L L E T I N

an overview of the current scenario of capacity building vis--vis the issues and future challenges,

M A R C H

2 0 1 3

EO missions, the CB requirement is also growing. NNRMS Standing Committee on Training (SC-T)

courses in geospatial technology (Tripathi, 2005). According to an estimate, as many as 1,50,000 students enrol for such courses either as a specialized course or as one of the subjects offered as a part of other courses. In some of the countries like Sweden, Australia, Germany, Canada, Switzerland, Denmark, USA, the subject is also taught in schools. A modest beginning has also been made in India and IIRS (Figure-1) has contributed in preparing the study material for schoollevel children. National Scenario and Role of ISRO in Capacity Building In India, over 90 universities/ institutes/ organizations offer the courses in this eld. While majority of the courses are offered at postgraduate level (Postgraduate Diploma/ M.Sc./M.Tech.), some of the universities do offer the courses at undergraduate level (B.Tech./ B.E./ B.Sc.). Additionally, a few institutes/ organizations, including the private owned ones, also offer Diploma/ Certicate/ Short-term training programmes as well. The mode of delivery of contents is primarily through regular class-room based programmes; however, of late a few universities/ institutes have also started distance learning programmes Department of Space (DOS) has recognized human and institutional capacity development in the area of space technology and its applications as a critical element primarily to harness the societal benets of its endeavours as well as promoting research in space science and technology. Capacity building programmes in the eld of EO technology and its applications including remote sensing for planetary exploration are conducted at its four centres Indian Institute of Remote Sensing (IIRS), Dehradun; National Remote Sensing Centre (NRSC), Hyderabad; Space Applications Centre (SAC), Ahmedabad; and Physical Research Laboratory (PRL), Ahmedabad. Recently, it has also set-up Indian Institute of Space Science and Technology (IIST) at Thiruvananthapuram, Asias rst Space Institute, to develop critical human resource required for the department for furthering the developments in space science and technology and achieving its long-term mission and goals. IIST is the rst institution in the world offering undergraduate, postgraduate and doctoral programmes in space science, technology and applications. ISRO/DOS also hosts and supports the Centre for Space Science and Technology Education in Asia and the Pacic (CSSTEAP), afliated to the United Nations, established during November, 2005 in response to the UN General Assembly Resolution (45/72 of 11.12.1990) endorsing the recommendations of UNISPACE-82, the United Nations Department for Outer Space Affairs. CSSTEAP is an educational and research institution in the eld of space science and technology. It offers Postgraduate level courses in the elds of RS& GIS, Satellite Communication, Satellite Meteorology, and Space Science with its course curricula developed by the United Nations. ISRO/DOS centres IIRS, SAC and PRL cater the needs of all its capacity building programmes. IIRS is the nodal agencyof ISRO/ DOS for capacity building in the eld of EO technology and its application and other allied disciplines of geospatial technology. The capacity building programmes, achievements and thrust areas are highlighted in the next section.

Capacity Building Programmes and Achievements at IIRS

Founded in 1966 (formerly known as Indian Photo-Interpretation Institute, IPI), IIRS is a premier institute in the country with a primary mandate to build capacity in Remote Sensing and Geoinformatics and their Applications through education and training programmes at postgraduate level. The Institute boasts to be the rst of its kind in entire South-East Asia and was established with the collaboration of the Govt. of the Netherlands. Realising the imminent need of focused efforts in capacity building to ensure efcient utilization of ISROs forthcoming EO systems in the areas of natural resource survey, Earth and atmospheric sciences and oceanography, IIRS has been given the status of a Unit of ISRO with effect from 30th April, 2011.

domains: Training & Education programmes; Research programmes; Outreach programmes

Training and Education Programmes The training and education programmes of the Institute are designed to meet the requirements of various target/user groups, i.e. for professionals at working, middle and supervisory levels, fresh graduates, researchers, academia, and decision makers. The duration of courses ranges from oneweek to two-years. The training programmes conducted by the Institute (Table-1) are broadly grouped into (1) Postgraduate Diploma programmes, (2) Certicate programmes (including NNRMS-ISRO sponsored programme for University faculty), (3) Awareness programmes, (4) Special on-demand/ tailor-made courses. The Postgraduate Diploma programmes are of 10 months duration; the Certicate Programmes are of 8 weeks duration; the Awareness programmes for Decision Makers are of 1 week duration; and Special courses are of 18 week duration. The education programmes conducted by the Institute include: (1) M.Tech. course of 24 months duration being conducted in collaboration with Andhra University, Visakhapatnam; and (2) M.Sc. course of 18 months duration being conducted in collaboration with the Faculty of Geoinformation Science & Earth Observation (ITC) of the University of Twente (UT), The Netherlands. In addition to the regular training and education programmes, the Institute also conducts satellite-based Distance Learning Programme since 2007 which is unique in the country complementing the education programmes of the Indian Universities. The capacity building programmes conducted by IIRS are listed in Table-1. In addition, the Institute also provides support to the Centre for Space Science and Technology Education in Asia and The Pacic (CSSTEAP), afliated to the United Nations, to conduct the RS & GIS training & education programmes at postgraduate level. The research and pilot projects carried out by the ofcer trainees and students of various courses are mainly focused towards exploring the potentials, including developing new methods and applications, of Indian EO data. Apart from regular programmes, special/ tailor-made courses are designed and conducted to explore the use of new sensors launched on Indian EO satellites and also based on the need of the stakeholder/ user departments. Special/ Tailor-Made Courses for User Ministries and Stakeholder Departments The demand for short, special/ tailor-made courses from user ministries and stakeholder departments has increased considerably during the last decade.These courses are especially designed to meet the objectives and requirements of the user organisations with regard to utilising EO data along with other allied geospatial technologies, viz. large scale mapping, digitization of land records, water resource development, disaster risk reduction, geological surveys, mineral exploration, environmental studies, infrastructure development, etc. Additionally, under the Joint Experiment Programme (JEP) and utilization programme of different IRS missions (e.g. IRS-UP, RISAT-JEP), capacity building programmes were also undertaken by IIRS in collaboration with other ISRO/DOS centres in order to prepare the users for utilization of different kinds of EO data.

149

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

The capacity building activities of the Institute are primarily grouped into the following three

Table 1: Capacity building programmes at IIRS in geospatial training and education. S.No. 1. Programme M.Tech. in Remote Sensing & GIS (afliated with Andhra University) Specialisations in Sustainable Agriculture; Forest Resources & Ecosystem Analysis; Geosciences; Human Settlement Analysis; Marine & Atmospheric Sciences; Water Resources; Satellite Image Analysis & Photogrammetry; Geoinformatics M.Sc. in Geo-Information science & Earth Observation (afliated with ITC, University of Twente, The Netherlands) Specialisations in Geoinformatics; Natural Hazards & Disaster Risk Management Postgraduate Diploma in Remote Sensing & GIS Specialisations in Sustainable Agriculture; Forest Resources & Ecosystem Analysis; Geosciences; Human Settlement Analysis; Marine & Atmospheric Sciences; Water Resources; Satellite Image Analysis & Photogrammetry Post-Graduate Diploma in Geo-Information Science & Earth Observation (with ITC, University of Twente, The Netherlands) Specialisations in Geoinformatics; Natural Hazards & Disaster Risk Management Certicate Course in Remote Sensing Remote Sensing & Image Analysis (for Indian User participants) International Programme Certicate Course in Remote Sensing, Geoinformatics (Sponsored by ITEC, Govt. of India) Remote Sensing (with emphasis on Digital Image Processing); Geoinformatics NNRMS-ISRO Sponsored Certicate Course in Remote Sensing & GIS For University Faculty Cartography & Mapping; GIS Technology & Advances; Agriculture & Soils; Forestry/ Ecology/ Environmental Sciences; Geosciences; Meteorology & Atmospheric Sciences; Urban & Regional Planning; Water Resources Awareness Programme Remote Sensing An Overview for Decision Makers Tailor-made on-demand Courses Satellite-Based Distance Learning Programmes Basics of RS, GIS & GPS; Advanced Courses on Specialized Themes Duration 24 months

2.

18 months

3.

10 months

4.

10 months

5. 6.

8 weeks 8 weeks

7.

8 weeks

8. 9. 10.

1 week 1 week to 8 weeks 8 weeks;2-4 weeks

For further details, refer the Institutes website (www.iirs.gov.in)

A few select user/ stakeholder departments for which the special/ tailor-made courses have been conducted by IIRS are listed in Figure-2. During the last 25 years, 2,817 professionals from various national
Fig. 2: Select user/ stakeholder departments beneted from special/ tailor-made training programmes during last 25 years (January 1988 to January 2013)

and international organisations, majority coming from the Indian user ministries and stakeholder

departments, have been trained. NNRMS Sponsored Programmes for University Faculty The Institute is playing a signicant role in conducting the ISRO-NNRMS sponsored training programmes for university faculty. The programme is of 8 weeks duration and is offered in eight specialisations in technology and application domains. The main objective of this programme is to train the university /college teachers in RS & GIS technology and applications, so that they can further teach students in their universities/colleges. Every year 50 to 60 faculty from different

programme is that many of the faculty who have undergone the training at IIRS have started RS & GIS courses in their universities/colleges. So far, IIRS has trained 716 faculty in this programme. International Programmes Indian Technical and Economic Co-operation (ITEC) and Special Commonwealth Africa Assistance Plan (SCAAP), Ministry of External Affairs, Govt. of India, have recognized IIRS for two short courses (8 weeks duration) on Remote Sensing with special emphasis on Digital Image Processing (DIP) (since 2000); Geoinformatics (since 2004). The main objective of this programme is to train the foreign nationals from developing countries in the eld of DIP and Geoinformatics. A total of 349 (+17 ongoing) professionals from various countries around the world have been beneted with this programme. IIRS/ ISRO has also made agreements with the Govt. of Venezuela and Govt. of Algeria under which officials from these countries are regularly nominated to undergo the training and education programmes at IIRS. Distance Learning Programmes To widen the outreach of capacity building, a state-of-art EDUSAT Studio (Figure-3) has been established at the Institute in the year 2007 for conducting distance learning programmes. The EDUSAT Studio,
Fig. 3: EDUSAT setup at IIRS

the rst of its kind in the country, is used to live-telecast the courses across the country

using the space communication through ISROs satellites, GSAT-3 and INSAT-4CR. IIRS has so far conducted 9 courses (10th in progress) 7 courses on Basics of Remote Sensing, Geographic Information System & Global Positioning System (8 weeks duration) and two short courses (3rd in progress) on advancedthemes such as Advances in Geoinformatics (2 weeks). At present, IIRS has established network with 72 universities/ institutes across the country and over 7500 participants have been beneted with this programme so far. E-Learning courses will also be launched in the current year to further enhance the outreach. CSSTEAP Programmes The Centre for Space Science and Technology Education in Asia and the Pacic (CSSTEAP), afliated to the United Nations, is located within the IIRS campus. As mentioned earlier, IIRS is the nodal centre for conducting the education and training programmes in RS & GIS, apart from providing the infrastructure and administrative support. The RS and GIS programmes include Postgraduate Diploma,M. Tech. (with Andhra University) and short courses on specialised themes. The course curriculum for Postgraduate Diploma and M.Tech.isdeveloped by the United Nations. So far, 16 Postgraduate Diploma 151

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

universities (and its afliated colleges) get trained at IIRS every year. An important outcome of this

courses (17th batch in progress) and 29 short-term courses/ workshops have been conducted. In Post-graduate Diploma courses,364 professionals from 34 countries and in 29 short courses/ workshops, 536 professionals from 33 countries have been trained. 113 participants have been awarded M.Tech degrees, so far. Outreach Programmes and Services In its endeavour to disseminate the knowledge in the eld of remote sensing and GIS and maintain linkages with the professionals in the eld and alumni, the Institute is actively engaged in the following outreach programmes: Supporting external students/trainees and researchers (Fellows of JRD Tata-INSA, IASc, DST, TWAS, and university students & researchers) for dissertation/ research Conducting classes/ briefing for UG/PG students visiting IIRS on study tour Linkage with alumni through CONTACT Newsletter Preparation and dissemination of learning materials and multi-media tutorials Hosting and supporting activities of Indian Society of Remote Sensing (ISRS) WebGIS Services on Biodiversity and Bio-resource Information Biodiversity Information System (www.bis.iirs.gov.in), and Indian Bio-resource Information Network (www.ibin.gov.in).

Achievements in Training and Education Programmes The Institute has trained 8685 professionals (till January, 2013), including 850 professionals from abroad representing over 90 countries mainly from the Asia, Africa and South America. Out of the total number of 8685, 6711 Table 2: Achievements in capacity building progress during last 25 years (January 1988 to January 2013) Programme Regular Programmes Special/ Tailor-made Programmes (including ITEC courses) Distance Learning Programmes CSSTEAP Programmes Professionals Trained About 4000 2817 7539 (from 73 universities/ institutes including 1 from Nepal) 634 from 34 countries

professionals have been trained during the last 25 years (January, 1988 to January, 2013).Further, over 7,500students/ researchers/ faculty from 72 universities/ institutes spread across the country have also beneted through satellite-based distance learning programmes being offered by the Institute since 2007. A summary is provided in Table-2. Research The research at the institute is an integral part of capacity building as the learning and the expertise developed is incorporated/ integrated in the education and training programmes. The Institute has a multi-disciplinary and problem oriented research agenda that focuses on technology development as well as land/ocean/atmosphere applications using EO data and geoinformatics tools and technologies. The Institute has proven its leadership in the country in developing key societal applications, like watershed management, ground water exploration, modelling urban dynamics, coastal zone management, irrigation water management, biodiversity characterisation, geo-hazards monitoring, assessment and modelling, to name a few. On the technology development side, commendable research in the areas of digital photogrammetry, microwave remote sensing, radar interferometry, advanced image processing techniques, LiDAR data processing, spatial data modelling, spatial decision support systems, etc. has/ is being carried out. Biodiversity Characterization at Landscape Level (BCLL) and National Carbon Project (NCP) are the benchmark studies in the last 25 years which were initiated at IIRS and have subsequently taken the shape of programmes at national level.

25 years include: (1) ISRO-Geosphere Biosphere Programme (ISRO-GBP); (2) Microwave Remote Sensing Programmes (JEPs/UPs); (3) Disaster Management Support Programme (DMSP); and (4) Earth Observation Applications Mission (EOAM). Enhanced focus is currently being made on microwave and hyperspectral remote sensing applications in diverse thematic areas and towards modeling and understanding Earths processes. Some of the broad areas of current research interest include: Microwave and Hyperspectral Remote Sensing; Satellite and Terrestrial Photogrammetry; Geospatial Data Modeling, Geo visualisation and WebGIS; Monitoring and Assessment of Himalayan Ecosystem Processes and Services; Measurement, Modeling and Assessment of Carbon Pools and Fluxes; Modeling Land use / Land cover (LULC) Dynamics; Hydrological Modeling including Assessing the Effect of LULC Dynamics and Climate Variability/ Change on Water Resources ; Mapping, Monitoring and Modeling of Geological Hazards and their Vulnerability and Risk Assessment; Erosion and Soil Nutrients Dynamics Modeling; Atmospheric and Climate Studies; Characterising Planetary Surfaces. An interdisciplinary research project on Monitoring and Assessment of Ecosystem Processes and Services in Western Himalaya has been initiated as the major research focus in 12th FYP. Using recent advances of EO techniques (focusing on ISRO missions) and allied spatial technologies supported by extensive eld investigation, eld instrumentation in several sub-themes, this study will address the major issues in the Western Himalaya, such as geodynamics and seismicity; quantication of erosional soil loss and its impact on crop productivity; biodiversity monitoring and modelling species and habitat loss and impact of nitrogen loading on ecosystem structure and function; snow-melt runoff and glacier mass balance studies; urban re-development planning; and study of extreme weather events. In addition, the institute has also contributed signicantly in national mission projects taken up by ISRO/DOS, such as LULC, Land Degradation, Geomorphological and Lineament mapping; Rajiv Gandhi National Drinking Water Mission (RGNDWM); Landslide Hazard Zonation (LHZ); National Urban Information System (NUIS) etc. In order to pursue the research agenda, state-of-art research facilities have also been established during the last 25 years. Instrumented Field Observatories Atmospheric CO2 Measurement Network (at Dehradun, Haldwani, Nainital, Gadanki and Mount Abu); Flux Towers for Measuring Energy, Water Vapour & CO2 Exchanges (three towers at Haldwani,Barkot and Saharanpur out of ve towers in the country) (Figure 4); Observatory for Aerosol Climatology (at Dehradun);

153

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

The major research programmes of ISRO in which the faculty has contributed in the last

 Field Observatory for Soil Erosion and Hydrological Modeling (near Dehradun and Roorkee). In-house Laboratories Computer Labs for Processing Remote Sensing, Photogrammetric & GIS Datasets; (Figure 5) Soil & Water Analysis Lab with precision instruments (Figure 6). Field Instrumentation A wide range of equipments for surveying and for measuring and monitoring physical,
Fig. 4: Flux tower at Barkot

chemical and environmental parameters.

Challenges and Thrust Areas in Capacity Building

Since the launch of IRS-1A in March 1988, Indias EO capability has grown manifold. The design and development of EO missions have primarily been guided by the NNRMS to meet the application requirements of different ministries of the Govt. of India towards food, water and environmental security, and disaster management. A large number of advanced satellites with improved spatial, spectral, temporal and radiometric resolutions are planned in future in order
Fig. 5: In-house computer laboratory

to ensure the continuity of EO services and meet the growing future requirements, particularly in the areas of meteorology and climate; ocean state forecasting; satellite based navigation for providing location-based services and real-time kinematics applications in land surveys; improved understanding of Earth system and its processes; and disaster management. In addition, a variety of space science and planetary exploration missions, including the human space ight mission, are planned to improve the scientic understanding of our solar system. The future EO infrastructure pose a challenge to the scientific community to developing improved methods and

Fig. 6: Soil & water analysis laboratory

techniques for processing the datasets

use the information generated from such systems for various societal/ sectoral applications and better understanding of the Earth system and its processes. Further, poor institutional and human capacity is recognized as one of the main issues in making many of the developmental programmes/ schemes of the Government fully successful; therefore, strong capacity building initiatives and efforts are thus needed to maximize the outreach of space-based inputs and services, both within the Government as well as in the community. Some of the thrust areas in capacity building in terms of technology and applications of future Indian EO missions are: Microwave remote sensing; Hyperspectral remote sensing; Satellite photogrammetry; LiDAR data processing; Large-scale/ cartographic mapping; GNSS applications including mobile applications in location-based services; Satellite meteorology; Geospatial modeling using open source tools and technologies; Data assimilation and coupling of EO data and GIS with numerical process-based models; Disaster mitigation and emergency management; Atmospheric and climate change studies; Geo-visualisation and Geo-web services; Volunteered geographic information for different applications; Planetary geosciences; Use of EO and other geospatial web portals (e.g. Bhuvan and India-WRIS) for local-level developmental planning. Further, in order to enhance the know-how to use EO and geospatial information and spreading awareness amongst the planning/ implementing agencies (ministries and stakeholder departments) and local people, the following thrust areas are also identied: Aligning the capacity building programmes in tune with developmental schemes of various ministries of the Government (e.g. Integrated Watershed Development Programme (IWDP), Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS), Accelerated Rural Water Supply Programme (ARWSP), Jawaharlal Nehru National Urban Renewal Mission (JNNURM), etc.) and requirements of geospatial industry; Strengthening the mechanism to share and disseminate geospatial data/ information at all levels including local people and establishing a strong network with other capacity building institutions including NGOs; Promoting web-enabled (e-learning) modular courses of shorter duration with credit system, virtual classrooms and virtual labs for off-campus students, linkage with foreign universities/ institutes and international experts in order to enhance the outreach and give geospatial education and training a global outlook; Expanding the target group from fresh graduates (present trend) to professionals and thematic/ domain experts within the Government functionaries, NGOs and other private institutions;

155

N N R M S

B U L L E T I N

M A R C H

2 0 1 3

obtained from such advanced systems, but even more importantly to the society so as to assimilate and

Standardisation of course curriculum and syllabus to maintain high quality and standards in geospatial education and training programmes conducted by various institutes/ organizations/ industry in the country, and accreditation of such courses/ programmes by the Indian university education system;

Enhanced research focus on developing new/ innovative methods and frameworks to promote interoperability and utilisation potential (linked to Government policies for socio-economic development) of upcoming EO systems through inter-departmental and international collaborations;

Including basic education about the concepts of EO technology and its potentials as a part of course curriculum at school and college level to bring awareness in the future generation; There is an urgent need to evolve a system to make local institutions/ people a part of the developmental process

in order to increase the acceptability and ownership of the Govt. programmes/ schemes. An initiative has been taken by ISRO/DOS in cooperation with State Remote Sensing Centres and stakeholder departments, in this direction through a project on Space-based Information Support for Decentralized Planning (SIS-DP), funded by the Planning Commission, Govt. of India. Under this initiative, geospatial data on different themes along with high-resolution (spatial) satellite images from Indian EO systems are not only being delivered at different levels of functionaries, including panchayat level, but also capacity building is undertaken to sensitise the Govt. ofcials and local people about the relevance and use of such geospatial data/ information in local-level developmental activities. However, a lot remains to be done to bring in the sense of appreciation of EO data and its potentials. As emphasized by the twelfth plan document of the Planning Commission, a strong hand-holding of the capacity building institutions and civil society and public-private partnership is especially needed for social mobilisation to harness the benets of geospatial information provided by the EO systems.

Conclusion
Human and institutional capacity building at all levels is the key for realizing full potential of geospatial data, information and services generated from Indias current and future EO missions. This article has highlighted the current global and national status of capacity building in EO technology and its applications, achievements of IIRS as a primary capacity building institute of ISRO, and challenges and thrust areas which need immediate attention, especially keeping in view the future EO missions. There is a need to standardize the curriculum and syllabus of geospatial education and training programmes conducted in the country and aligning them to suit the requirements of developmental schemes of various ministries and stakeholder departments of the Government and that of geospatial industry. The strong linkage and hand-holding of government/ nongovernment geospatial capacity building institutions with academia, NGOs and industry, supported with innovative methods of content delivery, research focus linked to developmental schemes of the government in addition to inter-departmental and international collaboration, are some of the important areas to facilitate the process of effective capacity building and bringing public awareness about the utilisation potential of EO data in National development.

Acknowledgements
The authors are grateful to the Chairman, ISRO and Secretary, DOS for his constant support, encouragement and guidance in capacity building endeavours of the institute (IIRS). Thanks to Director, Earth Observations System (EOS), ISRO Headquarters for his pursuance to write this article. The help rendered by Dr. Y.V.N. Krishna Murthy, Dr. S.K. Srivastav, Shri Prasun K. Gupta, Dr. Puneet Swaroop, Dr. S.P.S. Kushwaha and Dr. S.K. Sahain preparing the article is sincerely acknowledged.

References
Rhind, D. and Raper, J. (2004). Rethinking GIS education.http://www.gisdevelopment.net/ Education/papers/edpa0010pf. htm, accessed on February 25, 2007. Tripathi, N. (2005). The reality of Geoinformatics education in Asia. GIS Development - Asia Pacic, 2005. Spatial News, the leading GIS, LBS, and Geospatial technology news resource.[Online]. Available: http://spatialnews. geocomm.com/education/links/usa.html. [Accessed: 20-Feb-2013]. DRAFT Twelfth Five Year (2012-17). Planning Commission, YojanaBhawan, Government of India.[Online].Available: http://planningcommission.gov.in/plans/planrel/12thplan/welcome.html. [Accessed: 10-Feb-2013].