The GIS Awareness Booklet

This page reproduces the 15 page booklet accompanying the GeoData Institute's GIS Awareness Package, a set of resources produced for the UK academic community as part of the Information Technology Training Initiative (ITTI) in England.

The GIS Awareness booklet


GIS ... A New Experience For Your Data!

`Geographic Information Systems, usually abbreviated to GIS, is a term ... normally used to describe computer facilities, which are used to handle data referenced to the spatial domain, with the capability to inter-relate datasets, to carry out functions to assist in their analysis and the presentation of the results ...' Chorley Report, 1987.

Geographic Information Systems (GIS) have become a major IT development area in teaching, research, government, the public services, commerce and industry. The relative rapidity of this development means that the new technology is now involving many groups that had not previously been strongly committed to computer databases or computer graphics. There is therefore an immediate requirement to provide these people with an awareness of the concepts and capabilities of GIS followed by more in-depth training to provide the required skills.

The Information Technology Training Initiative (ITTI) launched by the Information Systems Committee of the University Funding Council provides a context for the development of awareness and support materials for use in both the academic and non-academic sectors. The materials are aimed mainly at:

This is the first in a series of products to promote the awareness of GIS and consists of a simple `Storyboard' - a non-interactive demonstration - plus this booklet. They aim to provide an initial general awareness of GIS principles and applications. Topics covered include:

What Is GIS ?

An understanding of what Geographic Information Systems represent may be helped by considering the component parts of the term separately.

Geographic...

This term is used because GIS tend to deal primarily with `geographic' or `spatial' features. These are objects which can be referenced or related to a specific location in space. The objects may be physical, cultural or economic in nature. Features on a map for instance are pictorial representations of spatial objects in the real world. Symbols, colours and line styles are used to represent the different spatial features on the two-dimensional map.

Computer technology has been able to assist in this mapping process through the development of automated cartography and computer aided design. Computer programs can now accomplish in minutes and hours tasks which previously took days or weeks for cartographers and draughtsmen to complete.

Information...

This represents the large volumes of data which are usually handled within a GIS. All real world objects have their own particular set of characteristics or descriptive attributes. This non-spatial alphanumeric data plus locational information needs to be stored and managed for all spatial features of interest.

Historically maintained as paper files, computer technology has enabled much more efficient handling and management of information within automated database management systems.

Systems...

This term is used to represent the systems approach taken by GIS, whereby complex environments are broken down into their component parts for ease of understanding and handling but are considered to form an integrated whole. Computer technology has aided and even necessitated this approach so that most information systems are now computer based.

Computer systems are becoming vital for the storage and manipulation of the increasing volumes of data, the handling of complex spatial algorithms and the integration of data of different scales, projections and formats. All of which are essential to GIS.

Geographic Information Systems...

Geographic Information Systems are thus usually computer based with an emphasis on preserving and utilising the inherent characteristics of spatial data, by handling both components of spatial data: the physical location in space and the set of characteristics associated with that location.

GIS tend to handle the two elements of spatial features separately, the spatial relationships being represented by graphical display and the attribute information being stored within a database. The GIS thus needs the ability to relate the attribute information to the spatial locality.

How Is The Real World Represented Within A GIS ?

The real world is far too complex to model in its entirety within any information system, so only specific areas of interest should be selected for inclusion within a given GIS application. Once a particular application area has been chosen the next task is to select those features which are relevant to the application and to capture information about their locations and characteristics. The GIS, being computer based, needs to have all of this information in digital form. It is thus necessary to consider how each real world feature can best be modelled within the computer system. There are essentially only five different types of spatial object, also known as entity, feature or facility, which can be represented within a GIS.

Point...

An object that occurs at one physical location in space and which has only one reference coordinate. Examples include trees, pylons, rainfall gauges, health clinics and hotels.

Line...

An object which spans between points and thus requires at least two reference coordinates, its start and end, to define its spatial location. Examples include roads, rivers, pipes and cables.

Area...

An object which has area and is defined by a continuous closed boundary. A number of coordinates are required to define its boundary. Area features are also known as polygons. Examples include fields, counties, lakes, planning sites, health districts and enumeration districts.

Surface...

A feature which requires three dimensions to define it. Thus a series of spatially distributed x,y coordinates are necessary to define a surface, each with a vertical z value. The z value may represent physical terrain, population density or rainfall, for example.

Network...

A feature defined by a series of line segments connected to form a continuous branching system of links. This structure enables the calculation of optimal routes through road networks or the simulation of flow through rivers or pipes.

It is the first three features which are the most commonly used but occasionally it is necessary to model more complex entities which require the use of surface or network facilities.

The above features can then be represented within a GIS in one of two quite different ways: vector or raster format.

Vector...

Positional data in the form of x,y coordinates. Each feature has a coordinate or string of coordinates to represent a particular location within a specific spatial referencing system. Spatial objects are thus defined by points and lines, in a similar way to conventional paper maps and drawings. Examples of data in vector format include site plans, ordnance survey maps and Computer-Aided Design (CAD) drawings.

Raster...

Data expressed as a matrix or array of grid cells or pixels. Each coordinate or value is represented by a cell in the regular array of cells. The position of spatial objects can thus only be defined to the nearest cell. Examples of data in raster format include scanned aerial photographs, satellite images and scanned documents or maps.

The two structures are each appropriate to different data sets and applications:

Vector
A vector structure can provide a flexible and accurate representation of an object due to the fine resolution obtainable with coordinate points. Vector structures also tend to incorporate the topology and other spatial relationships between the individual entities and are therefore ideally suited to representing linked networks such as pipe or road systems. It is very accurate for the measurement of areas or lengths and ideal where there is a requirement for cartographic-quality pen plots. Computer data storage is very economical but certain operations such as overlay analysis and proximity calculations have high computational requirements, which result either in slow operations or high hardware specification requirements. Manipulation and analysis of digital images, which are essentially raster, is not feasible.

Raster
A raster structure provides information at a much lower resolution since data can only be located to the nearest grid cell. Computer storage tends not to be economic, although data compression techniques are improving the situation. Operations such as overlay, buffering and neighbourhood analysis are, however, more efficiently accomplished with a raster structure. Raster structures are ideal where the source data is raster-based, such as satellite or scanned photogrammetric data, and particularly where the data also need to be output to a raster device.

Traditionally, commercial GIS were based on one format or the other and were not designed to handle both. Vector-based GIS tended to arise from CAD or automated cartography systems whilst raster systems grew from image processing technology. Most GIS software today will enable conversions between the two formats or will at least allow users to display vector data over the top of raster data, provided that the latter is geo-referenced first.

Layers...

Regardless of the way the data are structured, all GIS separate the different types of information into data `layers'. This means for instance that all the water features are held on one layer and all the roads on another. This allows for separate display and processing when necessary but does not prevent cross referencing between data layers during query and analysis. A number of data layers are thus built up into a sandwich within the GIS. Layers are referenced to a common spatial domain so that they can be scaled and overlain in such a way that any given reference point can be located on any of the layers and the data value extracted.

What Functionality Is Available Within A GIS ?

GIS have the ability to perform numerous tasks utilising both the spatial and attribute information stored within them. It is these often very sophisticated functions which are the real strength of a GIS. This functionality can be subdivided into four main groups.

Data Acquisition And Input...

This is the collection of data in both digital and analogue form and its transformation into an appropriate standardised format for entry into the GIS. Data sources include paper maps, satellite images, aerial photographs, field notes and other paper records. These need to be converted into a digital form, if they are not already, before the GIS can make use of them. Procedures such as digitising, scanning and manual keyboarding are involved in this conversion process.

Data Storage And Management...

The data are stored and indexed within a database system which facilitates shared access to the data, and which maintains the reliability, security and integrity of the data by controlling access to it and supervising updates. Database management thus tends to be at the heart of a GIS ensuring controlled and coordinated data retrieval and analysis. Ideally the data set should be structured in such a way as to be independent of the applications which access it.

Data Manipulation And Analysis...

Geographic analysis requires a close association between the spatial elements and their attribute data. Previously these have been held and managed separately within automated cartography packages and databases respectively. GIS provides the technology to perform more sophisticated analysis which makes use of the links between the two. Queries to a GIS can thus be graphics driven or data driven. Graphics-driven queries involve spatial based searches for objects and retrieval of the associated attribute data, or point and query functions to select displayed features and retrieve the associated information. Data-driven queries involve the use of data values selectively to display the matching spatial features or the use of attribute values to determine shading pattern or colour coding of the relevant spatial elements.

Operations to retrieve, summarise, selectively display and analyse both the alphanumeric and graphical data are usually embedded within a GIS. Examples of the types of spatial analysis usually available within a GIS include:

Choropleth mapping
Area features are classified according to their attributes. A legend acts as a look-up table with each range of attribute values being associated with a particular colour or shading pattern - eg. shading enumeration districts according to population density.

Buffer generation
Boundaries are created around points, lines or areas at an equal distance in all directions. The result being circular, corridor or polygon shaped features, respectively, which represent areas at set distances from the original object - eg. creation of buffers around an airport to calculate zones of different noise intensity.

Polygon overlay
Area features on one data layer are overlaid onto those of other data layers in order to calculate areas which have a certain combination of attributes or lack certain values - eg. overlaying soil, drainage and slope data layers will provide information on the land most suitable for crop production.

Contouring
The interpolation of lines of the same elevation from spot height data - eg. the calculation of height contours from which to generate slope and elevation data for land surfaces or any other statistical surfaces.

Terrain analysis
The creation of three-dimensional views or digital terrain models (DTM) from height data - eg. a three-dimensional view of a landscape can be used to assess the visual impact of an afforestation programme.

Network analysis
Tracing through a network of connected line features in order to simulate flows of traffic or water. Often used to calculate shortest or quickest paths between two points in a network - eg. use of a network to calculate optimum routes for emergency vehicles.

Area and length calculations
The calculation of the area of polygon features or the length of linear features. Often these simple calculations can be constrained to include only those features with attributes satisfying certain selection criteria - eg. calculation of the area of parkland within a city which needs to be maintained or calculation of the length of pipeline within a sewage network which needs replacing.

Examples of how the attribute information may be manipulated within a GIS include the retrieval of data values for selected features, the computation of attribute means and the calculation of statistical summaries and samples.

Data Presentation And Output...

Output from a GIS can be in the form of an interactive computer display of the graphic and/or attribute data or as traditional paper maps, tables, graphs and reports. Automated cartography techniques are employed to produce the graphical output from GIS. Commonly used output devices include pen plotters, ink-jet printers, electrostatic plotters, laser printers and screen copiers. Export to other systems such as statistical packages, word processors and multi-media will usually require the GIS to be able to convert internal data into other standard formats.

What Are The Benefits Of Using GIS ?

There are a number of advantages of implementing a particular application on a GIS, as opposed to developing it manually or on any other computer system. The first in the following list is the most tangible benefit and the reasoning many of the larger organisations give for investing in GIS technology. Further advantages of GIS however are less tangible and are therefore frequently left out of formal cost-benefit analyses but they are none the less important and in the long term are probably more important.

Existing Tasks More Efficiently...

A GIS has the major advantage of replacing paper maps and documents which have traditionally been duplicated throughout an organisation and which require frequent updating and replacing. Tangible cost benefits can thus be achieved by removing the need for such paper documents and converting to GIS technology. GIS based maps can quickly be updated, edited, printed or duplicated whereas traditional maps can take days of careful manual labour to achieve the same.

New Tasks Not Previously Possible...

A GIS provides the technology to perform tasks not previously possible because they were too time-consuming or not physically practical before automation. Multiple map overlays, generation of buffer corridors, surface interpolation and visualisation of terrain models are examples of such sophisticated analysis.

Data Management...

A GIS provides all the advantages of controlled information management such as sharing data between multiple users, reducing data duplication and increasing security, accuracy, integrity and validity of data. Shared access to a central database is much more efficient than providing numerous copies of the same data for everyone to view and alter as they wish. It is then that inconsistencies in the data arise and errors can develop and therefore the data become less reliable and useful.

Scenario Modelling...

A GIS provides capabilities to undertake modelling scenarios and test `what if ?' type queries. This is an extremely powerful tool for planners whereby different potential outcomes resulting from changes to the input parameters can be tested quickly and efficiently. The potential for better informed decision-making is thus greatly increased.

Value-Added Processing...

A GIS provides the potential to create new information from existing data, through selection and combination analysis techniques. Individual data layers can be combined in numerous ways to produce further information layers. The combination of slope, drainage, soils and vegetation data layers for instance could provide a graded map for erosion potential or crop suitability depending on the manner in which the various layers are combined.

Information Provision...

A GIS tends to be based around the realisation that information is a valuable resource to be utilised to its full potential. GIS provide very effective data management, retrieval and analysis tools, but perhaps the greatest strength of GIS is the capability to visualise spatial features and relationships. Knowledge of the location and characteristics of objects as well as their relationships to each other is crucial for effective management, planning and investment decisions. Increasingly it is becoming a statutory requirement for organisations to provide accurate and up to date information for a wide variety of purposes, thus the demand for GIS and other automated data handling systems is likely to increase.

What Are The Applications Of GIS ?

The applications or uses for a GIS are endless, wherever spatial features need to be modelled and analysed. Some common examples are listed below.

Environmental Resource Management...

Environmental applications lend themselves very well to GIS because they often require the integration of numerous different data sets during the analysis, since environmental systems tend to be complex and composed of inter-related sub-systems. Particular applications include river channel maintenance, coastal defence, forestry and national park management.

Emergency Planning And Routing...

The provision of optimum locations for emergency service centres can also be aided by GIS analysis of the various parameters such as access to roads, population density and various health indicators. Network analysis can be utilised to define optimum routes, such as shortest or fastest, for the routing of emergency service vehicles.

Provision Of Health, Educational Or Retail Services...

Consideration of the spatial distribution of different sectors of the population, their health and socio-economic characteristics and the accessibility to transport routes plus the location of existing facilities are required prior to the effective location of new facilities or the allocation of new services.

Facility Management For The Utilities...

The utility industries tend to have vast numbers of facilities to manage in order to provide large customer regions with an efficient and reliable service. Gas, water, electricity and sewage utilities for instance own a lot of land, buildings, cables, pipes and other physical facilities which need monitoring, maintaining and managing in order to provide an effective service.

Highway Maintenance And Accident Monitoring...

Roads and motorways need to be maintained and monitored for accident trouble spots. GIS are ideal for representing the spatial relationships between sections and storing the associated information tied to each section of road. Maintenance records can also be incorporated into the GIS and so provide up to date displays of the state of the road network and the sections which require immediate maintenance.

Market Analysis...

The spatial distribution of the population and particularly the different age groups and socio-economic sectors are essential information to the market analyst attempting to discover the most suitable place to launch a new product or sell a particular brand. The effectiveness of any given marketing stratey can also be modelled and evaluated.

Population Analysis And Prediction...

The spatial distribution of the population and the predicted level of a population are essential information to planners and developers when deciding what type of facilities need to be constructed now in order to best suit the needs of the future population. Census data thus provide an important input to GIS.

GIS essentially enable the relationships between various spatial features to be visualised and analysed which in turn encourages a better understanding of the interactions between the various features. GIS also enable the data to be manipulated and analysed quickly and flexibly in a single system which is an extremely powerful capability.

Other Important Issues To Consider

GIS are complex systems and can revolutionise the way organisations manage their information and approach certain spatially dominated applications, however the implementation of a GIS involves much more than hardware and software choices. There are five major groups of issues to consider.

Hardware...

The type of machines, number of terminals, networking requirements, digitising and printing needs all have to be considered for successful implementation of a GIS in an organisation. Final choices will depend on the budget available, the number and location of potential end users and the type of GIS to be installed.

Software...

The required operating system, database, GIS package and other supporting software also need to be decided. There is no `best' GIS software, the most appropriate choice depends on an individual organisation, its needs, the number and type of users, the type of applications involved and the budget at its disposal. GIS implementation can range from a PC based system for individual use, to departmental workstations and up to corporate systems based on a mainframe which serve an entire organisation. There is also the choice of a tool-box or menu-driven system. The former offers a high degree of flexibility in the implementation of a system so that it can be customised to an individual organisation's requirements and applications, however the effort involved in getting the GIS up and running is usually enormous. The latter choice is usually much more user-friendly and can be operational very quickly, however the choice of functions and interfaces is limited. Increasingly, however, commercial GIS will offer a basic menu-driven system but with the option of using a command or macro language to create tailored functions, interfaces or full applications.

Data...

The accuracy, source, ownership, copyright, confidentiality, security, standards and formats of data are important issues to consider. GIS technology allows the integration of data from a variety of sources, scales and formats for visualisation and analysis purposes, however the output from any GIS can only be as reliable and relevant as the information entered into it. Care is thus needed during data acquisition and input in order to maintain accurate and reliable data sets. This is a particularly important consideration since data acquisition and conversion usually represents the major component in the cost of implementing a GIS. Also as organisations become more open and begin to share their information it is vital to ensure the security and confidentiality within the database to safeguard the interests of the data owners.

Applications...

It is important to consider whether GIS technology is the most appropriate and relevant to any particular proposed application. Not all types of application benefit from GIS treatment. Projects which require the spatial analysis of a number of different data sets, require repeated access and query facilities, involve frequent updates and meet the demands of a number of personnel are valid for GIS implementation since the benefits of automation are enormous. However where an application is only relevant to a limited number of people, involves limited data layers and has only a one-off use, then the applicability of GIS technology is less than certain. Data acquisition and conversion tend to involve a considerable proportion of the total time taken to reach full GIS implementation and the costs involved with this stage may well not be worth it for the latter type of project.

Organisation...

The implementation of a GIS can have profound implications and ramifications within an organisation. In particular, corporate scale GIS will require previously disparate departments to cooperate and share information, will necessitate retraining of some staff and the employment of new ones, and will involve technology probably unfamiliar to most employees. Successful GIS implementation within an organisation appears to be dependent on a high level of staff awareness, involvement, training and support. The management should be supportive in this and provide sufficient time and resources. As with the incorporation of any IT development there will probably be resistance to change and a reluctance to assist from some of the staff. Experience seems to suggest that careful management is required at all stages of GIS implementation in order for the various organisational changes to occur smoothly.

GIS in Summary

A GIS is usually a computer-based system which provides facilities for data capture, storage, manipulation, analysis and presentation. The emphasis is on preserving and utilising the inherent characteristics of spatial data.

Spatial data comprise both a physical location in space plus a set of characteristics about that specific location. GIS recognise this fact and also that the different components of spatial information cannot be efficiently managed in the same manner. A graphical representation is most appropriate for visualising spatial relationships, whilst some kind of database is more appropriate for storing and analysing attribute information. A true GIS, therefore, needs the ability to relate the attribute information to the spatial locality.


GeoData Institute, GIS Awareness Booklet
jds@geodata.soton.ac.uk