Dolcine L1, Ahluwalia H1, Taylor T2, Brahm A1, Sharma P.K
1) Info-Electronics Systems Inc., (IES) Canada
2) PCI Geomatics, Canada
3) Punjab Remote Sensing Center, (PRSC) India
Agriculture has always been India’s most important economic sector. Large area yield forecasting prior to harvest is of interest to government agencies, commodity firms and producers. Timely monitoring during the growing season provides opportunities to mitigate any detected challenges. The current methods used in India provide neither sufficient nor timely cost effective spatial information. The purpose of this project is to develop an Agricultural GeoCapacity Centre Network to overcome these problems and provide a sustainable, long term solution. This presentation describes the functionalities of AGCN, the requirements, the design and the implementation AGCN is a complex system where several independent modules are integrated and cooperate to reach the final objective. It is based on a multi-tier architecture model. These modules are implemented as one or several applications; for crop monitoring, yield forecasting, weather forecast, etc. The applications are built around services provided by the GIS core, like Image Processing, Map Generation, Web System, and Database Connection. The GIS core provides application programming interface to build custom applications, integrates external application and access to the database management system.
Agriculture plays a critical role in rural employment and environment management (Philip et al, 2005) and will face many challenges over the next century. Sustainable land development practices are urgently needed all over the world to preserve the production potential of agricultural land while protecting the environment (FAO, 1993). An integrated approach for sustainable agricultural development requires a combination of agricultural status, surface and groundwater potential, soils, natural vegetation, terrain characteristics, wetland etc. with meteorological information and socio-economic conditions. According to different experts, within the next decade, conventional field surveys will be phased out due to prohibiting costs and likely to be replaced by tools of Information Technology and remote sensing (Bourrough, 1986) to meet the increased demand for up-to-date and reliable information.
Agriculture has always been India’s most important economic sector. Large area yield forecasting prior to harvest is of interest to government agencies, commodity firms and producers. Early information on yield and production helps in such activities as planning the type of crops (wheat, rice, barley, oats, etc.). Timely monitoring during the growing season provides opportunities to mitigate any revealed challenges. Information Technology, Remote Sensing Applications, and Decision Support Systems will facilitate decision makers’ and farmers’ knowledge of the state and situation of the agricultural system, thus helping them plan and act accordingly, Moni, 2003). Information and Communication Technology is a way to link the farmers to Universities and Research Centers, and researchers can orient their work to solve local problems in agriculture (Punjab Agricultural University, 1998).
Some forms of Agriculture Decision Support Systems have been discussed or implemented in different parts of the world. The Crop Condition Assessment Program (CCAP) ), (Stan Aronoff 2005) is a GIS based analytical Web application that shows in near real-time crop conditions within the Canadian prairies and the US. The MARS (Monitoring Agriculture with Remote Sensing) project (MARS Stat, 2004) focuses mainly on predicting the production volumes of the major crops at a national level and at a regional level for all EU member states. Their methodology is based on remote sensing, agro-meteorological data and crop growth model. The ‘Global Monitoring for Environment and Security’ (GMES) program is a European initiative and has a component which aims at supporting and complementing existing regional information and early warning systems on food and agriculture. The Global Information and Early Warning System on Food and Agriculture (GIEWS https://www.fao.org/giews/english/about.htm ) operated by the United Nations Food and Agriculture Organization (FAO) monitors famine-prone areas of the globe in order to predict food shortages and possible famine conditions. The Production Estimates and Crop Assessment Division (PECAD) (https://www.pecad.fas.usda.gov/) of the US Department of Agriculture (USDA) relies on several different data sources to monitor weather anomalies that affect crop production and quality of agricultural commodities (Tetrault and Baldwin,2006). The main agro-meteorological input data sources are the ground meteorological station measurements, and grid weather data integrated with satellite imagery.
In addition, several crop models and data reduction algorithms are using both the station and satellite agro-meteorological data sets. These models include crop calendars, crop hazards, and several different crop yield reduction models to assess crop conditions. In India, the concept of an Agricultural Resource Information System (AgRIS) has been extensively discussed (Moni, 2003). The author expects the AgRIS Project to emerge as the richest “agricultural resources information system” covering all aspects of agriculture, natural resources, and food systems, linking the farmers to Universities and Research Centres, whereby researchers can orient their studies to solve local problems in agriculture.
2 AN AGRICULTURAL GEOCAPACITY CENTER FOR INDIA
In order to develop an Agricultural GeoCapacity Center for India, the AGCN project has been initiated by an Indo-Canadian partnership. With the participation of the Canadian International Development Agency (CIDA), IES and PCI represent the Canadian team, while Punjab Remote Sensing Center (PRSC) and PAU (Punjab Agricultural University) represent the Indian counterpart.
Punjab, the home of the ‘Green Revolution’ in India, has been selected as a pilot area for the AGCN. After several discussion sessions organised in Ludhiana with PRSC and scientists at PAU, the following list of “burning issues” or main concerns for Punjab agriculture were identified:
- Declining water table and extensive use of rice-wheat cropping system.
- Declining soil health and increasing soil-water pollution.
- Improvement of agro-meteorological weather forecasts. Maximization of rainfall agriculture and crop diversification.
- Better use of rice-wheat straw and protection of environment.
- Improvements in soil-water-nutrient management.
- Integrated insect-pest-disease management.
- Enhancement in crop productivity, quality of produce and sustainability of productivity.
- Improvement of socio-economic status of Punjab farmers including issues of marketing, mechanization, land parcel size and others.
- Land use changes and diversification of cropping patterns/farming systems.
- Development of post-harvest technologies and value additions through processing.
- Development/training of scientific human resources.
Based on these requirements, AGCN has been designed as an ensemble of tools and applications to provide information for:
- Optimisation of water use for irrigation:
- Precise weather forecasts
- Maximization of rainfall agriculture and crop diversification
- Better planning of:
- Land use
- Farm input, storage and output
- Better modeling and analysis of the crop system based on a multitude of inputs including weather, soil characteristics, population needs, and other economical market parameters
- Guidelines and strategic decisions based on local geographical, industrial and socio-economical characteristics of the area .The proposed Agricultural GeoCapacity Center Network project entails the establishment of a nation-wide information technology system for collecting, collating, organizing and interpreting information on land and resources collection and managing socio-economic data.
The proposed AGCN system, as shown below, implements the above recommendations. It has enterprise architecture with core GeoCapacity services and data management. The enterprise architecture meets requirements, such as data distribution for efficient sharing and storage, services-oriented architecture for integrating best-of-breed technologies, multi-tier architecture for efficient and effective system enhancements, and environment componentization for system extensibility and scalability. The core of GeoCapacity services and data management provides fully automatic satellite data processing, centralized spatial data storage and management, and the interoperability required to access and use data and information from the various sources.
Figure 1 AGCN High-Level Architecture
The AGCN will provide optimized performance with a common platform for advanced geospatial analysis. A job control system will manage the creation, execution, and monitoring of the fully automated processes. Alternative tools and algorithms for image classification will also be provided. Users will be able to access services through the Intranet or Internet, and can select various processing algorithms (such as image registration, image mosaicking, data loading, and map generation) and output formats (such as Web-based or printed maps). Also, various agriculture-specific applications can easily be built on top of the GeoCapacity core and services.
The AGCN is a modular system. Based on user gathered requirements, a number of modules have been identified within AGCN: Weather Application Module, Water Management Module, Cropping System Analysis Module, Soil Management Module, Pest and Disease Module, as well as Socio-Economic Module. The modules will be implemented as one or several applications for crop monitoring, crop yield forecasting, weather forecasting, etc. The applications are built around services provided mainly by the GIS core, like Image Processing Service, Map Generation Service, Web System Service, and Database Connection Service. The GIS core is the heart of the AGCN upon which everything is built. The GIS core provides services to build custom applications, and GIS desktop provides access to the database management system. The database management system will store, modify, and extract geospatial and non-geospatial information from the database.
3 AGCN APPLICATION MODULES
The next section briefly discusses the main components of the system: Weather applications, crop system and water management modules.
3.1 Weather Application Module
The weather application module consists of the following parts:
- Acquisition, checks and processing of daily meteorological station data
- Spatial interpolation
- Production of weather indicators and daily Web-based maps
- Production of meteorological inputs for the statistical and dynamical crop models
- Processing of the land use map and management practices data and use of water budget and crop water requirements models, production of seasonal map products for drought analysis and water deficit.
Figure 2 Weather Application Modules
Daily meteorological station data and weather forecasts are used in different ways for crop yield evaluations. The information provided by this module is first used as weather indicators for direct evaluation of alarming situations such as droughts, extreme rainfall during different crop development phases of sowing, flowering and harvest; the second as input for the crop growth model and statistical yield model (MARS stat, 2004). The main objectives pursued with the weather module are:
- To evaluate the effects of weather on crops yields
- To produce inputs for the crop growth model for the monitoring and simulation of the crop behaviors.
3.2 Crop System Module
The crops’ behaviors are mainly influenced by the atmospheric conditions near the earth’s surface. The main purpose of the crop system module is to provide:
- Seasonal crop yield forecasts using statistical and dynamical crop models.
- Cropping patterns and crop rotation maps.
- Indices to characterize long term changes in the cropping system. Various indices, such as Multiple Cropping Index (MCI), Area Diversity Index (ADI), and Cultivated Land Utilization Index (CLUI) will be generated.
Figure 3 Crop Application Modules
The cropping system analysis module produces estimated crop indicators like yield and production maps, cropping pattern maps and indices on performance of the cropping patterns. Different activities are expected in this module, some of the tasks like model validation will be done once at the start of the project and every five years:
- Collection and processing of input data
- Verification and validation of statistical model and dynamical models for yield forecast.
- Methodology for crop acreage estimation
- Spatial schematization (at grid level or agro-climatic zone)
- Crop simulation
- Spatial aggregation and production of crop indicators maps
The module uses daily interpolated grid weather to simulate biomass accumulation, crop development and yield forecast. Besides crop monitoring, the module will produce crop indicator maps and generate alarm warnings in case of abnormal conditions. The crop growth model is a complex process which takes place on farms at field level. Crop yields vary among regions, farms, fields, and years. The cropping system analysis takes into account the influence of factors like soil and weather, the influence of some other factors are omitted or considered constant.
3.3 Water Management Module
This module is responsible for the computation of monthly and annual water balance; this water balance is the difference between potential evapotranspiration and rainfall. The water resource of Punjab is estimated as the sum of surface and ground water; a water resource map will be produced. For the monitoring of the ground water level and quality, different maps based sample data from wells and piezometers will also be developed. For the day-to-day and long-term management of the water resources, an ensemble of models will be integrated to interact with the AGCN core system. These models include ground water model, hydrologic and flood forecasting models, reservoir model, hydraulic and flood mapping models. This module is developed in a companion paper entitled “Implementation of an Integrated Decision Support System for Water Management in Souss-Massa Morocco”.
Figure 4 Water Management Module
4 AGCN SYSTEM ARCHITECTURE
Figure 1 above illustrates the five parts of the overall AGCN system from a high-level perspective:
- Agriculture Data Input includes communication, digitization and entry of relevant agriculture data (from weather to water, soil and statistics). The source of data could be either digitized disk file data or on-line/in-situ data transmitted through the network.
- Application-Specific Services are high level services that are specifically responsible for handling application-dependent data access and logics/models. Examples are from Weather Services to Water Services to Yield Forecast Models.
- Core Geo-Capacity Center Services and Data Management provides the core generic spatial (raster and vector) data processing services, spatial data access services, and spatial data storage and management.
- Applications are specific agriculture applications built on top of both the Application-Specific Services and the Geo-Capacity Center. Examples are from Wheat Yield Forecast to Water Table Monitoring and Post-Harvest Planning/Management.
- Products are results generated by the Applications. They could be either maps or reports or both.
To build the entire AGCN system, enterprise system architecture will be used. Enterprise system architecture meets the main system requirements as data centralization for efficient access and management, data distribution for sharing and storage efficiency, services oriented architecture for best-of-breed technologies integration, multi-tier architecture for efficient & effective system enhancement, and componentization environment for system extensibility and scalability. The available enterprise technology and relevant information technology (IT) are matured enough for implementation of the AGCN system and also ensure the security and performance requirements.
The inclusion of the Geo-Capacity Center in the AGCN system turns the enterprise system into a Geo-Capacity Information System (GCIS), which provides the ability to handle spatial raster/vector data. The spatial-enabled ability allows the AGCN system to practicably solve real-world agriculture problems, such as spatial-based monitoring, forecasting and management. The main technologies required for implementing the Core Geo-Capacity Center Services and Data Management are as follows:
- Image processing technology. Abundant libraries of various image processing algorithms exist nowadays. The maturity of these algorithms has been proved by many applications for decades.
- Componentization environment for the image processing technology. Recent technology advancement in pluggable function framework provides a componentization environment for applying mature processing algorithms.
- Application Programming Interface (API). Application Programming Interface and Software Development Kit provide an environment for users to develop their own algorithms in the pluggable function framework which an off-the-shelf system does not provide. The new pluggable functions developed can be easily deployed into the componentization environment.
- Spatial Database. Spatial database provides an efficient data structure (objectrelational table) to store spatial raster/vector data and optimize query performance.
- Geo-Database Management (GDM) technology. GDM technology handles different formats of raster/vector data and works seamlessly with spatial databases.
- OGC Web Services. OGC (Open Geospatial Consortium) provides international standards for publishing raster and vector data for sharing among different data providers. These standards and technology provide relevant products and services (Web Feature/Map/Coverage Services) for data publication, query and retrieval.
- RAID Storage System. The current RAID hardware storage system can hold huge amounts of data, up to Terabytes of data, to meet the data storage requirement for the AGCN system. More importantly, the RAID system provides maximum reading/writing speed and reliable data safety for the AGCN system.
To develop high-level Application-Specific Services for the AGCN system, Application programming interface and software development kit technology can be used to develop the corresponding pluggable functions and componentization environment required by the AGCN system.
The development of the specific Applications could be diversified using different tools. Maintenance of a standard communication protocol (such as HTTP and/or SOAP) between the specific applications and the backend services (including the core Geo- Capacity services) is required.
The Products are usually presented in a map format. The core Geo-Capacity services provides components/tools for data analysis and map generation.
Figure 5 blueprints the system architecture for AGCN. It depicts an implementation of the overall architecture shown in Figure 1 with components specified with respect to the requirements of the different applications.
The application/technical view shown in Figure 5 covers the major components and their inter-connection in a multi-tier enterprise framework for the applications:
- Data tier. This tier is responsible for storing and managing the agriculturerelated spatial data. A spatial database is recommended. Multiple spatial database servers can work as a grid system or as remotely distributed systems.
- Services tier. This is responsible for providing services to access and process data, conduct agriculture-relevant analysis, and disseminate user-required information. It can be divided into two sub-tiers:
- Data access sub-tier. This tier will provide a standard interface to access data from either homogeneous or heterogeneous database systems. Upon availability, the data can be registered and published for access from permitted services.
- Services sub-tier. This tier consists of various agriculture-related components to provide services from data processing to monitoring, prediction and management.
- Presentation tier. This tier contains agriculture specific applications which integrate the available services and provide user interfaces for their clients.
Figure 5 AGCN System Architecture
The AGCN for the implementation of an Agriculture Information system is based on the Geo-Capacity Information System (GCIS) concept and is conceived with the aim of developing methods to produce timely statistics on land use, planted area and production volumes for various crops within India, of applying remote sensing and ground surveys to estimate the planted area, as well as providing weather information and water management service applications and tools. Real-time image processing tools, proven methods to relate satellite imagery to quantitative crop yields, weather, soil and crop information to feed crop growth monitoring and water models, information technology required to build an open and multi-tier architecture are presently available, such that implementation and deployment of an AGCN system is not beyond our reach.
References FAO (1993): World Soil Resource, Report 73, Food and Agriculture Organisation, Rome. MARS Stat, 2004: Methodology of the MARS Crop Yield Forecasting System. Volume 4. Also available on CD-Rom and the Internet: . Stan A ( 2005): ‘Remote Sensing for GIS Managers’, ESRI Press: 341-342. Burrough P.A (1986): Principle of Geographic Information System for Land Use Assessment.
PAU (1998): Punjab Agriculture University, Vision 2020. Moni M. (2003): Impact of economic reforms on Indian Agricultural Sector: Application of geomatics technology to reduce marginalization and vulnerability of small farmers in Indian. Agriculture Informatics Division, National Informatics Centre (NIC).
Available at . Tetrault B. and Baldwin B. (2006): Monitoring Global Crop Condition Indicators Using a Web-Based Visualization Tool. Monitoring Science and Technology Symposium: Unifying Knowledge for Sustainability in the Western Hemisphere Proceedings RMRSP- 42CD. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. p. 744-748.