Geographic Information Systems
in Sustainable Development

Courtesy :   Geographic Information Systems Group

                      Environment and Natural Resources Service (SDRN)
                      FAO Research, Extension and Training Division 

Introduction

The importance of this integrated approach to development and management of natural resources have been emphasised in many international fora on sustainable development. The 1992 United Nations Conference on Environment and Development (UNCED) devoted Chapter 10 of its Agenda 21 to this topic, noting that: 

"...Expanding human requirements and economic activities are placing ever increasing pressures on land resources, creating competition and conflicts and resulting in suboptimal use of both land and land resources. If, in the future, human requirements are to be met in a sustainable manner, it is now essential to resolve these conflicts and move towards more effective and efficient use of land and its natural resources. Integrated physical and landuse planning and management is an eminently practical way to achieve this. By examining all uses of land in an integrated manner, it makes it possible to minimise conflicts, to make the most efficient tradeoffs and to link social and economic development with environmental protection and enhancement, thus helping to achieve the objectives of sustainable development. The essence of the integrated approach finds expression in the coordination of the sectoral planning and management activities concerned with the various aspects of land use and land resources." 

This Special provides a general introduction to Geographic Information System technology, explores some of its most common applications (including FAO's use of GIS), and supplies a page of links to digital datasets available worldwide.

1. Background: FAO and natural resources databases
More than any other agency within the United Nations system, the Food and Agriculture Organization (FAO) deals continually with issues of development and management of renewable natural resources. Soil, water, climate, plants, animals, people - and the ways in which they interact - are at the heart of the FAO mandate to assist developing countries to increase food production and to provide food security. FAO is a source of global, regional and subregional information on land-use planning and management, and provides technical assistance in these areas to member countries. 

Meeting these responsibilities requires the collection, analysis and dissemination of massive amounts of different kinds of information: about soils, rainfall, vegetation and land uses; locations of towns, highways, railroads and waterways; figures on population, income, health and nutrition, just to name a few. 
In general, information in GIS maps are of two types: 

• attribute data (e.g. statistics or text such as slope, soil type, vegetative cover type, etc.) contained in tables 

• spatial information, contained in the lines, points or polygons on maps. 

The resource planner's problem is how to compare and combine selected information from different maps, in order to evaluate a given geographic location or to assess the status of one particular area in relation to other areas (for example, which of two areas would be more suitable for a given crop). 

The figure below illustrates the spatial and the attribute information found in GIS maps, in this case of the Bay of Bengal. The attribute information consists of two items: length-of-growing-period (LGP) zones (in days), and the major climates. The spatial information is contained in the boundaries of the LGP zones and the major climates are designated by different colours/shades on the map. For every polygon (closed area) on the map, there is one record (row) in the attribute database table. What makes GIS different from other kinds of computer mapping systems is that these two kinds of information are always linked and processed jointly in GIS. 

Computer technology has long made it possible to manipulate and analyse statistical information. But only recently has technology been developed that can convert maps into a computer-usable digital format and allow the simultaneous manipulation of both the geographic spatial data and related attribute data. This makes it possible for the resource planner to rapidly produce a combination of maps and tables that answer such questions as: 

• What is the present and potential fuelwood production within X kilometer of Town A? 

• Where should a given food-processing facility be located? 

• What raw materials or markets are found within 300 km of the site? What and where are the available transport and power systems? 

• What will be the effect on farm income in zone H if the amount of land devoted to growing crop 1 is reduced by 20 percent over the next biennium? 

• What are the locations and sizes of zones within a certain administrative unit that satisfy a given set of conditions, such as: 

• soil type A or B; 

• amount of rainfall between C and D mm; 

• summer temperature between X and Y degrees; 

• population density and income distribution at E and F respectively; and 

• distance to roads less than G km? 

As pressures on land and water resources continue to mount, the ability to accurately assess resource conditions and trends becomes daily more essential. A Geographic Information System (GIS) is a powerful information tool at the disposal of decision-makers.

2. Concepts, history and comparison with other tools
Basic concepts of GIS

The term GIS is currently applied to computerized information storage, processing and retrieval systems that have hardware and software specifically designed to cope with geographically referenced spatial data and corresponding attribute information. The spatial data is commonly in the form of "layers" (see figure below), which may depict topography, water availability, soil types, forests and grasslands, climate, geology, population, landownership, administrative boundaries and infrastructure (highways, railroads, electricity or communications systems).

Its capability of combining different map layers - in an operation known as "overlaying" - is one of the most important GIS functions. For a description of the overlay function, follow this link 

Short history of GIS
During the 1960s and 1970s, new trends arose in the way maps were being used to assess resources and to plan land use. Realizing that different aspects of the earth's surface did not function independently of each other, people recognized the need to evaluate them in an integrated, multidisciplinary way. One way of doing this was to simply overlay transparent copies of resource maps on light tables and look for places where the various attributes on the maps coincided. 

This technique was later adapted to emerging computer technology. In what was known as a grid-cell or raster system, simple maps were made by printing statistical values on a grid of plain paper, overlaying the grid values and using the overprinting of lineprinter characters to produce grey scales representing the statistical values. The results of these early methods, however, were not refined enough to find acceptance among cartographers. 

By the late 1970s, however, the technology of computer map-making had advanced rapidly, with literally hundreds of computer systems and programs being developed for various cartographic applications. At the same time, advances were being made in a number of related fields, including soil science, surveying, photogrammetry and remote sensing. The rapid pace of this development initially resulted in much duplication of effort in the various related disciplines. But as systems multiplied and experience was gained, the potential for linking different kinds of spatial data processing together into truly general-purpose geographic information systems emerged. 

By the early 1980s, as computer technology became more sophisticated, less expensive and more widely adopted, GIS came into its own. Today, GIS systems are rapidly being established by public agencies, research laboratories, academic institutions, private industry and public utilities. 

Comparison of GIS and with other computer systems
In essence, GIS is a data base management system (DBMS) specifically designed for simultaneous processing of spatial and related attribute data. A DBMS generally provides a language for analyzing data which allows users to describe to the system what they want to know with little or no attention to the mechanics or methods used by the system. A DBMS must also contain procedures for checking consistency of the data and maintaining their integrity. 

In addition to DBMS, GIS also has many capabilities similar to automated map making, computer-assisted cartography and computer graphics systems. However, as well as having a powerful capability for processing graphics, GIS must also be able to process non-graphic attributes, such as statistical data, in conjunction with the spatial data to which they are related. For example, if the user modifies the spatial data the GIS will make the necessary modifications in the related statistical database automatically. This link between the two types of data must be present if a system is to be considered a true GIS. 

Although GIS differs from other tools, such as tabular data base management systems, computer graphics, and automated map making, each of these other systems is, in fact, a component of GIS. What GIS does is make it possible to integrate all of them in one operation. 

One of the primary sources of geographic data used in GIS is information about the earth that is obtained through remote sensing. Remote sensing data are usually acquired either as digital satellite imagery or aerial photographs. After these images are geometrically corrected, enhanced, analysed and interpreted, the results can be fed into the GIS and integrated with other geographic data bases. 

The main components of GIS
Geographic Information Systems have three major components: computer hardware, sets of software, and the human resources and organization that make the system work. 

a) GIS computer hardware
The hardware components of a GIS include units that are common to any computerized data base management system - a general purpose computer, several disk drive units for storing data and programs, tape drives for back up copies of data, colour graphic display units, and other general purpose computer peripherals. 

The GIS has, in addition, several specialized hardware components, including: a digitizer or scanner, which is used to convert the geographical information from maps into digital form and send it to the computer; a plotter, which prints out the maps and other graphic outputs of the system; and a visual colour graphics workstation on which spatial data editing and display can be performed by the user. 

b) GIS software
The main GIS software components are designed to perform the following functions, where data implies both cartographic and/or attribute data: 

• data input: digitizing or scanning the lines on the maps and entering the attribute information from a keyboard 

• data base management 

• data analysis and processing 

• interaction with the user (map editing) 

• data output and presentation (plotting) 

Data input involves the conversion of data from maps, field observations, processed satellite images and aerial photographs into compatible digital form. 

Many GISs today utilize a manual digitizing approach to input maps. This means that someone must sit down with the map at a large, flat, digitizing table, and using a small cursor pad, follow the thousands of lines that make up the map, carefully keeping the cursor (cross hairs) on the lines, ensuring that lines are not double digitized or left out, and that intersections are accurately closed and no gaps are left in lines. 

However, large cartographic data inputs are generally made using automated digitizing systems such as scanners. These eliminate the manual work of following the lines and ensure consistent, repeatable results each time a map is scanned. Although scanning is quicker than digitizing, only good quality maps can be scanned, and even then the quality of the products is generally not as high. However, as in most areas of computerization, the technology is continually being improved. Furthermore, once a map has been digitized, it can be reproduced and transformed at will (much as a written document can be quickly edited or corrected once it has been entered into a word processor). 

The quality of input data will affect the quality of GIS products regardless of the sophistication of its hardware and software. In many cases, inventories of natural resources are often not completed or up to date and information in maps may have to be revised before digitizing. 

Data base management operations mainly consist of the following functions: structure, query, analysis and reporting of the attribute data linked to the features on the maps. 

Data output and presentation deals with the way the information is displayed to the user. This can either be as a visual display (soft copy) or hard copy drawn by a plotter, or as magnetically recorded or printed information in digital form. The plotter is to the GIS what a printer is to the standard word processor: it produces a copy of map on paper. 

c) Human resources and organization

When describing a GIS one tends to think in terms of hardware and software as the entire system, which overlooks perhaps the most important component: the people needed to make the whole system function effectively. 

It may seem that GIS is the resource planner's crystal ball, but - as with any computer system - the information produced is only as good as the information that is put in. Incorrect or inadequate information fed into the GIS will produce incorrect or inadequate answers, no matter how refined or "user-friendly" the computer technology may be. 

As in any map-making operation, data collection and data input operations require high standards of design and work, intensive training and frequent monitoring for quality control. In other words, in addition to having the right hardware and software to do the job, effective utilization of a GIS requires adequate staff training as well as planning, organization and supervision in order to maintain the quality of the data and the integrity of the final product. 

Another essential element of successful GIS operation is the need for data input and processing to be a joint effort involving the computer specialist and the subject matter specialist (e.g. crop production, forest management, aquaculture). This ensures that the necessary specialized subject matter expertise is applied in the interpretation and evaluation of data. Specialists in remote sensing and cartography may also be involved.

 
In many developing countries, resource information collection and processing systems are still relatively undeveloped. This means that application of GIS at the country and subcountry level will, in many cases, need to be accompanied by the improvement of existing information collection systems and the introduction of new ones. This provides an opportunity for international assistance, and imposes on FAO and other technical assistance agencies an added reason to develop their own capabilities in GIS and related technologies in order to provide technical expertise at the national level. 

Data formats or models: Vectors and rasters
Geographic Information Systems solve the problem of graphically representing a map in two basic ways, either as a raster or a vector form. 

In a vector-based system, the line work is represented by a set of straight-line segments called vectors. The X,Y coordinates at the end of each vector segment are digitized and explicitly stored, and the connections are implied through the organization of the points in the database	In a raster or cell-based system, the map is represented by a geometric array of rectangular or square cells, each with an assigned value.

4. Applications of GIS technology

An easy way to think of how GIS can be applied is to think in terms of the questions that the user might want answers to. As has been mentioned, one of the first steps when setting up a GIS is to survey the potential users to determine their information needs, and to identify those needs that can best be met by a GIS incorporating various combinations of data retrieval and transformation. 

The ultimate use of GIS lies in its capability for modeling: constructing models of the real world from digital data bases, and using these models to simulate the effect of a specific process over time for a given scenario. Modeling is a powerful tool for analysing trends and identifying factors that affect them, or for displaying the possible consequences of planning decisions or projects that affect resource use and management. 

At the continental level, for example, terrain maps can be combined with hydrologic maps and climatological data to produce maps of land suitability for various types or intensities of use, or specific crops. Demographic and administrative data can be added to provide projections of future supply-and-demand scenarios by region or country. 

At the national and local level, possible GIS applications are almost endless. For example, to decide on the best potential sites for growing a certain cash crop, the agricultural planner might use geographic data bases combining soils, topography and rainfall to determine the size and location of biologically suitable areas, and then overlay this with landownership and transport infrastructure, labour availability and distance to market centres. Further, he or she could then change the characteristics of various attributes over time to determine the probable impacts of changing circumstances, such as the effects of a drought, the rise or fall of domestic or world prices, or the development of additional roads. 

Applications of GIS to fisheries can take many forms. However, it is convenient to categorize them first of all as applications in capture fisheries and in aquaculture. For capture fisheries, GIS can deal with the spatial aspects of the three main fishery "realms", both individually and collectively - the environment, the fishery resources and the fisheries. GIS, using information from a variety of sources, including passive and active remote sensing, can predict where the fish will be, can be used for management, control and surveillance (e.g. monitor fishing) and can optimize fishing operations such as trade-offs between distance to fishing grounds and markets.

In aquaculture, GIS has been used to forecast development prospects using suite of parameters that vary geographically . Basically, this kind of application reduces to two broad questions: a) what is the suitability of any given area for the culture system (e.g., soil suitability for the construction of fish ponds) and b) what is the suitability of an area for fish growth (e.g. favourable temperature regime). Another GIS application is for the management of expanding aquaculture in the context of other, competing uses of land and water. The pertinent question here is: How much is too much?

Below is a map produced by an FAO inland fisheries GIS project.

Forestry planners can use GIS to monitor the impacts of deforestation, and to plan the timing and type of timber management practices based on information on soil types, species requirements, growth and yield, and even to assess the visual impacts of timber harvesting in sensitive scenic areas. 

The wildlife manager can use GIS to determine the size and location of animal populations, to map supply-and-demand relationships to meet consumption needs, or to determine areas having high food and habitat potential for specific species. 

In summary, what the GIS provides is a means of converting spatial data into digital form that can then be displayed, manipulated, modified and analysed and reproduced quickly in a new format, available for either visual display or hard copy reproduction. Conventional (paper) maps, in contrast, are time-consuming to prepare manually, and the display and analysis of changed data or the comparison of more than one set of map data (soil and vegetation, for example) requires additional manual labour. 

The digital data can also be easily transmitted from one user to another or from one GIS to another merely on disk, tape or by the Internet. As digital maps come into wider use, the cost of digitizing can be shared by many users. In fact, some digitized maps on CD-ROMs cost less than the same maps on paper. As networks and libraries of databases grow, information exchange should reduce the need for redigitizing regional or national maps and other geographic databases than are in common use. 

Courtesy - FAO, Rome