Introduction to GIS

AN INTRODUCTION TO

GEOGRAPHICAL
INFORMATION
SYSTEMS

IAN HEYWOOD
SARAH CORNELIUS

STEVE CARVER

THIRD EDITION

AN INTRODUCTION TO

GEOGRAPHICAL
INFORMATION
SYSTEMS

IAN HEYWOOD
SARAH CORNELIUS

STEVE CARVER

THIRD EDITION

AN INTRODUCTION TO
GEOGRAPHICAL INFORMATION SYSTEMS

Visit the An Introduction to Geographical Information Systems,
Third Edition, Companion Website at
www.pearsoned.co.uk/heywood to find valuable
student learning material including:

• Activities and revision questions for self-study or discussion
• Multiple choice questions to test understanding
• Weblinks for further investigation
• Datasets offering practice opportunities

We work with leading authors to develop the
strongest educational materials in geography,
bringing cutting-edge thinking and best
learning practice to a global market.

Under a range of well-known imprints, including
Prentice Hall we craft high quality print and
electronic publications which help readers to understand
and apply their content, whether studying or at work.

To find out more about the complete range of our
publishing, please visit us on the World Wide Web at:
www.pearsoned.co.uk

IAN HEYWOOD SARAH CORNELIUS STEVE CARVER

AN INTRODUCTION TO

GEOGRAPHICAL
INFORMATION SYSTEMS

THIRD EDITION

Pearson Education Limited
Edinburgh Gate
Harlow
Essex CM20 2JE
England

and Associated Companies throughout the world

Visit us on the World Wide Web at:
www.pearsoned.co.uk

First published 1998
Second edition published 2002
Third edition published 2006

© Pearson Education Limited 1998, 2002, 2006

The rights of Ian Heywood, Sarah Cornelius and Steve Carver to be identified as authors of this
work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without either the prior written permission of the
publisher or a licence permitting restricted copying in the United Kingdom issued by the
Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP.

All trademarks used herein are the property of their respective owners. The use of
any trademark in this text does not vest in the author or publisher any trademark
ownership rights in such trademarks, nor does the use of such trademarks imply any
affiliation with or endorsement of this book by such owners.

ISBN: 978-0-13-129317-5

British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data
Heywood, D. Ian

An introduction to geographical information systems / Ian Heywood, Sarah Cornelius,
Steve Carver.--3rd ed.

p. cm.
Includes bibliographical references and index.
ISBN 0-13-129317-6 (pbk)
1. Geographic information systems. I. Cornelius, Sarah. II. Carver, Steve. III. Title.
G70.212.H49 2006
910.285--dc22

2005056378

10 9 8 7 6 5 4 3 2
10 09 08 07

Typeset in 11/13pt Spectrum by 30
Printed and bound by Mateu-Cromo, Artes Graficas, Spain

The publisher’s policy is to use paper manufactured from sustainable forests.

Brief contents

List of figures x
Preface xv
Guided tour xx
Case studies xxii
Acknowledgements xxiv
Publisher’s acknowledgements xxvi
Abbreviations and acronyms xxviii
About the authors xxx

G Part 1 Fundamentals of GIS

1 What is GIS? 2
2 Spatial data 31
3 Spatial data modelling 71
4 Database management 108
5 Data input and editing 133
6 Data analysis 170
7 Analytical modelling in GIS 221
8 Output: from new maps to enhanced decisions 252

G Part 2 Issues in GIS

9 The development of computer methods for handling spatial data 280
10 Data quality issues 299
11 Human and organizational issues 333
12 GIS project design and management 355
13 The future of GIS 376

References 400
Glossary 412
Index 420



Contents Modelling the fourth dimension 101
Conclusions 105
List of figures x Revision questions 106
Preface xv Further study – activities 106
Guided tour xx Further study – reading 106
Case studies xxii Web links 107
Acknowledgements xxiv
Publisher’s acknowledgements xxvi 4 Database management 108
Abbreviations and acronyms xxviii
About the authors xxx Learning outcomes 108 111
Introduction 109
G Part 1 Fundamentals of GIS Why choose a database approach?
Database data models 114
1 What is GIS? 2 Creating a database 116
GIS database applications 120
Learning outcomes 2 Developments in databases 128
Introduction 3 Conclusions 130
Defining GIS 18 Revision questions 131
Components of a GIS 19 Further study – activities 131
Conclusions 28 Further study – reading 131
Revision questions 28 Web links 132
Further study – activities 29
Further study – reading 29 5 Data input and editing 133
Web links 30
Learning outcomes 133 164
2 Spatial data 31 Introduction 134
Methods of data input 135
Learning outcomes 31 Data editing 151
Introduction 32 Towards an integrated database
Maps and their influence on the Conclusions 168
Revision questions 168
character of spatial data 35 Further study – activities 168
Thematic characteristics of Further study – reading 168
Web links 169
spatial data 52
Other sources of spatial data 53 6 Data analysis 170
Conclusions 67
Revision questions 68 Learning outcomes 170
Further study – activities 68 Introduction 171
Further study – reading 68 Measurements in GIS – lengths,
Web links 69
perimeters and areas 172
3 Spatial data modelling 71 Queries 175
Reclassification 176
Learning outcomes 71 Buffering and neighbourhood
Introduction 72
Entity definition 74 functions 177
Spatial data models 77 Integrating data – map overlay 184
Spatial data structures 79 Spatial interpolation 194
Modelling surfaces 87 Analysis of surfaces 202
Modelling networks 94 Network analysis 213
Building computer worlds 97
Modelling the third dimension 99

viii Contents

Conclusions 217 Conclusions 296
Revision questions 218 Revision questions 296
Further study – activities 218 Further study – activities 297
Further study – reading 219 Further study – reading 297
Web links 220 Web links 298

7 Analytical modelling in GIS 221 10 Data quality issues 299 300

Learning outcomes 221 Learning outcomes 299
Introduction 222 Introduction 300
Process models 222 Describing data quality and errors
Modelling physical and Sources of error in GIS 305
Finding and modelling errors in
environmental processes 226
Modelling human processes 234 GIS 318
Modelling the decision-making Managing GIS error 324
Conclusions 326
process 238 Revision questions 330
Problems with using GIS to model Further study – activities 331
Further study – reading 331
spatial processes 246 Web links 332
Conclusions 249
Revision questions 249 11 Human and organizational
Further study – activities 249 issues 333
Further study – reading 250
Web links 251 Learning outcomes 333
Introduction 334
8 Output: from new maps to 272 GIS applications 334
enhanced decisions 252 GIS users 338
Justifying the investment in GIS 343
Learning outcomes 252 Choosing and implementing a GIS 345
Introduction 253 Organizational changes due to GIS 351
Maps as output 254 Conclusions 352
Non-cartographic output 264 Revision questions 353
Spatial multimedia 265 Further study – activities 353
Mechanisms of delivery 266 Further study – reading 353
GIS and spatial decision support Web links 354
Conclusions 276
Revision questions 277 12 GIS project design and
Further study – activities 277 management 355
Further study – reading 277
Web links 278 Learning outcomes 355
Introduction 356
G Part 2 Issues in GIS Problem identification 356
Designing a data model 359
9 The development of computer Project management 365
methods for handling Implementation problems 371
spatial data 280 Project evaluation 371
Conclusions 373
Learning outcomes 280 Revision questions 374
Introduction 281 Further study – activities 374
Handling spatial data manually 281 Further study – reading 374
The development of computer methods Web links 375

for handling spatial data 283
The development of GIS 287

Contents ix

13 The future of GIS 376 Revision questions 398
Further study – activities 398
Learning outcomes 376 Further study – reading 398
Introduction 377 Web links 399
GIS in the 1990s 377
Where next for GIS in the twenty-first References 400
Glossary 412
century? 385 Index 420
Conclusions 390
Epilogue 392

Supporting resources

Visit www.pearsoned.co.uk/heywood to find valuable online resources

Companion Website for students
• Activities and revision questions for self-study or discussion
• Multiple choice questions to test understanding
• Weblinks for further investigation
• Hypothetical datasets offering practice opportunities in ESRI-

and MapInfo-readable formats and in Excel

For instructors
• Instructor’s Manual including lecture outlines and practical exercises
• PowerPoint slides containing the artwork and images from the text

Also: The Companion Website provides the following features:

• Search tool to help locate specific items of content
• E-mail results and profile tools to send results of quizzes to instructors
• Online help and support to assist with website usage and troubleshooting

For more information please contact your local Pearson Education sales
representative or visit www.pearsoned.co.uk/heywood

List of figures

1.1 Facilities in Happy Valley ski resort 2.13 Projections: cylindrical; azimuthal; conic
1.2 Avalanche! 2.14 Latitude and longitude
1.3 A nuclear power station and nuclear waste 2.15 Latitude and longitude of Moscow:

containers calculating the latitude; calculating the
1.4 Anti-nuclear protestors longitude
1.5 Tracing paper overlay 2.16 The Ordnance Survey National Grid system
1.6 Using GIS for siting a NIREX waste site 2.17 The UK postcode system
1.7 Radioactive waste case study: geology, 2.18 Marlin GIS at Network Rail
2.19 Aerial photographs
population, transport and conservation 2.20 Varying scale on aerial
criteria 2.21 Examples of satellite imagery; SPOT; Landsat
1.8 Radioactive waste case study: results from TM; MSS; Meteosat; IKONOS satellite
different siting scenarios imagery
1.9 Zdarske Vrchy 2.22 LiDAR imagery
1.10 GIS landscape assessment data layers: Zdarske 2.23 Map of hydrological connectivity in Upper
Vrchy Wharfedale, northern England
1.11 Using GIS for identifying conservation zones 2.24 Topographic Index: 2 metre; 16 metre; 64
in Zdarske Vrchy metre resolution DEM
1.12 Using GIS to assist in house hunting 2.25 Area of catchment with Topographic Index
1.13 The house hunting case study affected by grips
1.14 GIS workstations: dedicated GIS workstation, 2.26 Modern field survey equipment: EDM; laser
desktop GIS and GIS on hand-held devices range finder in use
1.15 The GIS interface: command line; GUI 2.27 GPS receiver and satellite configuration
1.16 Weather station and ski piste in Happy Valley
1.17 Points, lines and areas 3.1 Stages involved in constructing a GIS data
1.18 Examples of raster and vector GIS data layers model
1.19 Mapping, querying and modelling soil
information 3.2 A simple spatial entity model for Happy
1.20 From topographic maps to data layers Valley
1.21 People are a key component of GIS
3.3 Examples of surface data: terrain; rainfall;
2.1 Primary data sources for the Happy Valley population
2.2 Secondary data sources for the Happy Valley
2.3 Avalanche incident report, 14 February 2002 3.4 Elevation and snow depth in Happy Valley
2.4 Examples of paper maps and maps in use 3.5 Examples of network data in New Zealand:
2.5 Cold war propaganda map
2.6 Expressions of scale railways; rivers; roads
2.7 Real-world objects commonly stored as a point 3.6 Roads, rivers and sewage networks in Happy
2.8 Real-world objects stored as lines
2.9 Real-world objects commonly represented as Valley
3.7 Woodland types and entity definition problems
an area 3.8 An engineer’s view of a ski lift
2.10 Representing a city at different map scales 3.9 Raster and vector spatial data
2.11 Scale-related generalization 3.10 Effect of changing resolution in the vector
2.12 The Earth from space and some commonly
and raster worlds
used global map projections 3.11 A simple raster data structure: entity model;

cell values; file structure
3.12 Feature coding of cells in the raster world:

entity model; cell values; file structure

List of figures xi

3.13 Raster data compaction techniques: run 4.8 Relational tables for Happy Valley database

length encoding; block encoding; chain 4.9 Linking spatial and attribute data in GIS

encoding 4.10 Overlay of planning scheme, cadastral plan,

3.14 The quadtree existing roads and encroachments within the

3.15 Data structures in the vector world: simple East Legon Ambassadorial Area

data structure; point dictionary 4.11 Coded encroachments on the CP, showing

3.16 Topological structuring of complex areas also the boundary of the VOR

3.17 Ordnance Survey MasterMap: topography; 4.12 Client–Server Web GIS

imagery; ITN; address layer 4.13 A networked Web

3.18 Example surface types: 3D contour; wire 4.14 The general query interface in WICID prior

frame; shaded relief; 3D DEM; air photo drape to selection

3.19 Contours and spot heights 4.15 Summary of query

3.20 Stereoscopic satellite imagery and derived 4.16 On-screen preview of flows extracted

DTM data 4.17 Object hierarchy for Happy Valley OO

3.21 Examples of SRTM, SAR and LiDAR data database

3.22 Using differential GPS to check and rectify

DTM data 5.1 The data stream

3.23 Raster DTM: simple terrain; complex terrain 5.2 Digitizing table and PC workstation

3.24 Digital terrain models: vector grid; vector TIN 5.3 Digitizing software

3.25 Example DEM and TIN model for region of 5.4 Point and stream mode digitizing

varying complexity 5.5 Bézier curves and splines

3.26 Network data model 5.6 On-screen digitizing

3.27 Examples of GIS networks: road; river 5.7 Types of scanner: feed roller; flat bed

3.28 Link, turn and stop impedances affecting the 5.8 Colour separates: original; cyan; magenta;

journey of a delivery van yellow; black

3.29 Two representations of the London 5.9 Infrared QuickBird imagery used for

Underground network: linear cartogram; real mapping burned areas

space 5.10 Fire and fuel break in Pinus pinea stand as

3.30 The layer-based approach seen on the ground and in panchromatic

3.31 The object-oriented approach QuickBird image

3.32 A wire frame perspective of a terrain model 5.11 Examples of spatial error in vector data

3.33 Examples of true 3D data structures: 3D 5.12 Examples of original data problems and the

geological; CAD corrected data after processing

3.34 Far Point, Scolt Head Island, Norfolk (eastern 5.13 Radius Topology Feature Snapping

England). Snapshot contour maps 5.14 Filtering noise from a raster data set

3.35 Far Point, Scolt Head Island, Norfolk (eastern 5.15 Topological mismatch between data in

England). Difference map different projections

5.16 The results of repeated line thinning

4.1 Database facilities in GIS 5.17 Edge matching

4.2 The traditional approach to data handling 5.18 Rubber sheeting

4.3 Card index record from ski school manual 5.19 Data collection workflow

database 5.20 River Aire footbridge on GMS2

4.4 The database approach to data handling 5.21 Updated digital data showing footbridge and

4.5 Relational database table data for Happy layers

Valley 5.22 Surveying the footbridge

4.6 Database terminology applied to Happy 5.23 Zdarske Vrchy case study: forest cover,

Valley table created by reclassifying a LandstatTM scene

4.7 Happy Valley EAM diagram for the area

xii List of figures

6.1 Raster GIS measurements: Pythagorean 6.27 Example of cartographic hillshading

distance; Manhattan distance; proximity techniques: ‘Mont Blanc’ (section); ‘Die

distances; and perimeter and area Alpen’ (section); 3D hillshade view

6.2 House hunting case study: distance from 6.28 Modelling incoming solar radiation in Happy

office calculated using proximity method Valley representing morning to evening

6.3 Calculating lengths and areas from the during a winter’s day

Happy Valley GIS 6.29 Slope curvature and feature extraction:

6.4 Vector GIS measurements: distance and area profile curvature; plan curvature; feature

6.5 Boolean operators: Venn diagrams extraction

6.6 Buffer zones around point, line and area 6.30 Three scales of DTM feature classification of

features the northern Lake District using convexity

6.7 House hunting case study: distance from measures to characterize shape

office adjusted for road network 6.31 Summits identified by relative drop from a

6.8 Radioactive waste case study: 3 km buffer DTM of the northern Lake District

zones around the rail network 6.32 Ray tracing for visibility analysis

6.9 Proximity map for hotels in Happy Valley: 6.33 Results of viewshed analyses for single point

distance surface; 125 m buffer zones and linear feature

6.10 Raster GIS filter operations 6.34 The three capes area near Portland in south-

6.11 DEM of The Rocky Mountains, Colorado western Victoria and Cape Bridgewater: the

6.12 Results of remoteness model compared to site of 7 proposed turbines

simple linear distance: remoteness; linear 6.35 Degree of visibility for the single turbine on

distance; residuals Cape Bridgewater

6.13 Vector overlays: point-in-polygon; line-in- 6.36 A comparison of the visibility of the seven

polygon; polygon-on-polygon Cape Bridgewater turbines based on light of

6.14 Point-in-polygon analysis: simple case; sight alone

complex case; problem cases 6.37 Network data analysis: allocation of ski

6.15 Identifying areas suitable for a nuclear waste school to clients on a network

repository

6.16 Radioactive waste case study: results of vector 7.1 Natural analogue model for predicting

overlay showing the intersect of the rail avalanche hazard

buffer and clay geology 7.2 Simplified conceptual model of avalanche

6.17 Raster overlays: point-in-polygon (using prediction

add); line-in-polygon (using add); polygon- 7.3 Regression model of slope angle against

on-polygon (using add); polygon-on-polygon avalanche size

(Boolean alternatives) 7.4 Methodology for estimating emissions within

6.18 Original terrain surface with sample points a GIS framework

6.19 Four different interpolation methods applied 7.5 Estimates of the emissions of NOx

to sampled terrain data 7.6 A simplified conceptual forest fire model

6.20 Constructing a Thiessen polygon net 7.7 Derivation of catchment variables using a

6.21 Fitting a trend surface DEM

6.22 Spatial moving average in vector and raster GIS 7.8 Residuals (percentage difference between

6.23 Error maps showing interpolation errors for actual and predicted sales) for The Speciality

the four methods applied to sampled terrain Depot’s store network in Toronto, Ontario,

data in Figures 6.18 and 6.19 Canada

6.24 Example of 3D terrain rendering 7.9 Applying a simple linear weighted

6.25 Visualizing a new ski piste in Happy Valley summation model in raster GIS

6.26 Three-dimensional analysis: shaded slope and 7.10 Weighting data layers in the house hunting

aspect image: altitude; slope; aspect case study

List of figures xiii

7.11 User inputted area of perceived ‘High crime’ 10.8 Problems with remotely sensed imagery

together with attribute data 10.9 Land use change detection using remotely

7.12 Total crime densities for Leeds for all crimes sensed imagery

committed in 2002 10.10 Digitizing errors

7.13 Severe soil erosion on agricultural land 10.11 Topological errors in vector GIS: effects of

7.14 Forest fire in the Canadian Rockies tolerances on topological cleaning;

7.15 Example of a model builder interface (Idrisi32 topological ambiguities in raster to vector

Macro Modeler) conversion

10.12 Vector to raster classification error

8.1 Examples of GIS output from the Happy 10.13 Topological errors in vector GIS: loss of

Valley GIS connectivity and creation of false

8.2 Map components connectivity; loss of information

8.3 Example map 10.14 Effect of grid orientation and origin on

8.4 Cartographic symbolism rasterization

8.5 Percentage of council housing by census ward 10.15 Generation of sliver polygons
8.6 Example area cartogram: unemployment by 10.16 Point-in-polygon categories of containment
10.17 Simulating effects of DEM error and
population
algorithm uncertainty on derived stream
8.7 3D view of the retail trade areas for two stores
networks
operating within the same market
10.18 Simulating the effects of DEM error and
8.8 Example of linked displays in GIS
algorithm uncertainty on radio
8.9 Example of multimedia content in GIS displays
communications in the Happy Valley area
8.10 The Virtual Field Course
10.19 Resampling SRTM data from 100 m to 20 km
8.11 Layering a Pollution Surface (Nitrogen
resolution
Dioxide NOx) onto Virtual London
10.20 DEM error and ice sheet modelling
8.12 Different Renditions of Virtual London
10.21 Simulating uncertainty in the siting of
8.13 PPGIS integrated with a geo-referenced nuclear waste facilities
threaded discussion forum
10.22 Bootstrapping or ‘leave one out’ analysis
8.14 Example of poor quality output

9.1 Land capability mapping 11.1 Development of GIS applications
9.2 Examples of SYMAP and SYMVU output 11.2 GIS users
9.3 Land use and land capability for agriculture 11.3 Cost–benefit graph

9.4 Example of GBF/DIME data for Montgomery 12.1 Rich picture for the house-hunting GIS
County, Maryland
12.2 A simple map algebra equation

10.1 Accuracy versus precision 12.3 Example of part of the analysis required by

10.2 Resolution and generalization of raster the house-hunting GIS

datasets 12.4 The system life cycle ‘waterfall’ model

12.5 The prototyping approach
10.3 Scale-related generalization: 1:10,000; 1:50,000; 12.6 GANTT chart for the house-hunting GIS
1:500,000
12.7 PERT chart for the house-hunting GIS

10.4 Two mental maps of the location of

Northallerton in the UK 13.1 Example 3D object modelling in MapInfo
10.5 The Finsteraarhorn, Switzerland (4273 m) using Encom’s Engage3D add-on
10.6 Terrain model of Mount Everest and its
13.2 Moore’s Law
surrounding area based on photogrammetric 13.3 In-car navigation system

survey data 13.4 Google Earth

10.7 Multiple representations of Tryfan, north 13.5 GAM output showing locations of probable

Wales disease clusters

xiv List of figures

13.6 Example real-time traffic map, Houston, 13.10 ESRI’s Hurricane Disaster viewer
Texas 13.11 Before and after images of areas hit by 2004

13.7 Mapped flooding depths in New Orleans after Boxing Day tsunami
Hurricane Katrina 13.12 Inundation models for predicted global sea

13.8 Predicted weather patterns in Hurricane level rise
Katrina 13.13 Monitoring ice conditions in the Antarctic
13.14 Deforestation in the Amazon Basin
13.9 Hurricane Katrina, before and after satellite
images of Biloxi

Preface

I HOW THIS BOOK IS WRITTEN Therefore, we concentrate on those areas that
enable you to make sense of the application of GIS,
Your encounters with GIS to date may be similar to understand the theories upon which it is based and
those of a Martian arriving on Earth and being faced appreciate how to set up and implement your own
with a motor car. Imagine a Martian coming to GIS project.
Earth and stumbling across a motor car showroom.
Very soon he (or she) has heard of a ‘car’ and may I THE THIRD EDITION
even have seen a few glossy brochures. Perhaps you
are in the same position. You have heard of the term The third edition has been expanded to cover the
GIS, maybe even seen one or two demonstrations or latest advances in GIS, incorporate new specially
the paper output they produce. commissioned case studies demonstrating GIS in
action and take advantage of the opportunities
Developing the analogy of the Martian and the offered by a move to full colour presentation.
car leads us to a dilemma. There are two approaches
to explaining to the Martian what the car is and how The case studies are written by academics, profes-
it works. The first method is a bottom-up approach. sionals, teachers and research students to help
This involves taking the car apart into its compo- illustrate the many ways in which GIS is being used
nent pieces and explaining what each part does. around the world. Using colour throughout the
Gradually, we put the pieces back together so that book we have been able to include better examples
by the time we have reassembled the car we have a of GIS output as well as photographs and screen
good appreciation of how the car works in theory. shots that help you explore applications and case
However, we may still have little idea about how to studies in more depth. New reflection questions and
use it or what to do with it in practice. activities help you check your understanding of
important concepts.
The second method, the top-down approach,
starts by providing several examples of what the car Many of the changes and corrections have been
is used for and why. Perhaps we take it for a test run, made in response to suggestions made by reviewers,
and then explore how the different components of and from lecturers and teachers using the book.
the car work together to produce an end result. If Many, many thanks to all of you who commented
this approach is adopted, we may never be able to on the first and second editions.
build a car engine, but we will have a clear apprecia-
tion of how, when, why and where a car could be I HOW TO USE THIS BOOK
used. In addition, if we explore the subject in suffi-
cient technical detail we will know how to choose There are three ways to use this book:
one car in preference to another or when to switch
on the lights rather than the windscreen wipers. 1 Browse through the text, consider the chapter
headings, and review the learning outcomes at
We feel that the same two methods can be used to the start of each chapter. Supplemented by
inform you about GIS. Since we believe you are read- reading Chapter 1 this will give you a quick tour
ing this book not because you want to write your through the world of GIS and highlight some of
own GIS software, but because you wish to develop a the important issues and concepts. If you wish
better appreciation of GIS, the approach adopted is you can explore specific topics further using the
similar to the top-down method. We focus on the boxes, case studies and the companion website.
practical application of GIS technology and where
necessary and appropriate take a more detailed look 2 A better way to approach the text is to read the
at how it works. In a book of this size it is impossible book chapter by chapter. This will allow you to
for us to explain and describe every aspect of GIS.

xvi Preface

discover the ideas presented and develop an familiarity with PC computing. Anyone who has any
understanding at your own pace. To assist you we experience of word processing, spreadsheets, data-
have included learning outcomes at the start of bases or mapping packages, should be able to apply
each chapter, reflection boxes within each that knowledge in a GIS context. The typical first-
chapter, and revision questions and pointers for year undergraduate IT course offered in subject
further study at the end of each chapter. The areas like geography, biology, business studies or
reflection boxes can be found at the end of major geology is sufficient for you to cope with the ideas
sections of text to encourage you to pause and presented in this book. We assume that you are
reflect on key issues and themes. They may ask familiar with terms such as hardware and software,
you to think beyond the text itself and about the and the major components of a computer: for
role of GIS in your studies, your work and the example, monitor, keyboard, hard disk drive, CD-
world around you. The revision questions are ROM drive, processor and memory. We make no
provided to encourage you to examine and revisit other assumptions – this book is written to be acces-
the major themes and issues introduced in each sible to students of GIS from any professional or
chapter and reinforce your learning. Further academic background: from archaeology, through
study sections at the end of each chapter have biology, business studies, computing, demography,
two parts; suggestions for further reading environmental management, forestry, geography,
including full references and bibliographic details, history … and on to zoology.
and suggestions for activities you can do yourself
either on the web, on paper or using a GIS If you want to become a GIS expert, you will need
package if you have one available. A selection of to be comfortable with more advanced computing
web links is also provided at the end of each issues, and will have to expand your computing
chapter to help you explore online materials and background to include skills in such areas as pro-
information resources. gramming and networks. These issues are beyond
the scope of this book, but we hope to provide you
3 The third way to use this book is as a reference with a valuable head start and many pointers as to
source. You might wish to delve into those where to continue your journey.
sections that you feel are relevant at a particular
time. Perhaps you can read the appropriate A book cannot substitute for hands-on experience
chapter to supplement a course you are in a subject area as practical as GIS. Therefore, for a
following. There is a comprehensive index and fuller appreciation of GIS we encourage you to
glossary at the back of the text and pointers to enroll on a course that offers practical experience of
further sources of information. We have tried to GIS, or to find a system to use in your own time.
reference only readily available published There are a number of excellent GIS ‘courses’ now
material. References to other additional available electronically and reference has been made
information sources on the Internet can be found to these on the website for the book.
on the website for the book. These lists of further
sources are by no means comprehensive, but are I HOW THIS BOOK IS STRUCTURED
offered as a starting point. We suggest that you
also consult library and online sources for more The book is organized into 13 chapters in two parts.
up-to-date materials. The first part of the book deals with GIS theory and
concepts. After an introductory scene-setting chap-
I BEFORE YOU START ter, which also introduces four case studies which
are used throughout the book, important topics like
You could not drive a car without an understanding spatial data, database theory, analysis operations and
of the road network or road signs. Similarly, a full output are all examined. In the second part of the
understanding of GIS requires some computing book we have focused on a selection of GIS ‘issues’ –
background, particularly in topics like operating sys- the development of GIS, data quality, organizational
tems and file management. This book assumes basic issues and project management. There are, of
course, other issues we could have considered, but

Preface xvii

those selected reflect many of the key themes of A review of how the traditional map-making process
current interest, and areas important for anyone shapes these characteristics is presented. The three
undertaking practical work in GIS. The book ends basic spatial entity types (points, lines and areas),
with a chapter on the future of GIS in which a which form the basic building blocks of most GIS
reflective look is taken at current GIS, and some pre- applications, are introduced. Maps and a range of
dictions are made for the future of the technology. other sources of spatial data are reviewed.
Each chapter has been carefully structured to link
together into a unified whole, but can be read in iso- Chapter 3: Spatial data modelling
lation. There are several common elements to each
chapter to help you get the most from the book. How do you model spatial form in the computer?
These are: This chapter considers in detail how the basic spatial
entities (points, lines and areas) can be represented
I Introduction and list of learning outcomes. using two different approaches: raster and vector.
I Boxes providing additional detail on theory and Two other entity types that allow the modelling of
more complex spatial features (networks and sur-
practice. faces) are introduced. Finally, modelling of three-
I Full colour maps, diagrams, tables and and four-dimensional spatial data is reviewed.

photographs. Chapter 4: Database management
I Illustrated case studies with further details of
This chapter introduces the methods available for
issues and applications in GIS. handling attribute data in GIS. The need for formal
I Reflection questions and activities. methods for database management is discussed. The
I Conclusions including revision questions. principles and implementation of a relational database
I Pointers to further study including references to model are considered in detail, since this model is the
most frequently used in current GIS. Database options
additional reading and activities. for large-scale users are presented, including the use
I Web links. of centralized and distributed database systems.
Finally, a brief introduction to the object-oriented
In summary, the chapters are: approach to database management is provided.

I PART 1: FUNDAMENTALS OF GIS

Chapter 1: What is GIS? Chapter 5: Data input and editing

This chapter provides an overview of GIS. It exam- This chapter gives an overview of the process of creat-
ines what GIS is, what it can do and, in brief, how it ing an integrated GIS. It takes us from source data
works. The chapter starts by looking at the types of through data encoding, to editing and on to manipula-
generic questions GIS can answer and expands on tory operations such as re-projection, transformation,
these with reference to a series of case studies which and rubber sheeting. The chapter provides examples of
are then used throughout the rest of the book. GIS how these processes are carried out, and highlights
is then defined, and a range of issues and ideas asso- issues pertinent to the successful implementation of a
ciated with its use identified. Much of the material GIS application.
introduced in this chapter will be covered in more
detail later in the book. Chapter 6: Data analysis

Chapter 2: Spatial data Methods for making measurements and performing
queries in GIS are introduced in this chapter.
This chapter looks at the distinction between data Proximity, neighbourhood and reclassification func-
and information and identifies the three main tions are outlined, then methods for integrating data
dimensions of data (temporal, thematic and spatial). using overlay functions explained. Interpolation
The main characteristics of spatial data are described.

xviii Preface

techniques (used for the prediction of data at present a comprehensive history of GIS but aims to
unknown locations) are introduced and the analysis give some context for the systems and concepts we
of surfaces and networks considered. Finally, analysis work with today.
of quantitative data is reviewed.
Chapter 10: Data quality issues
Chapter 7: Analytical modelling in GIS
The terms used for data errors and quality are
This chapter provides a summary of process models explained at the beginning of this chapter, since the
before considering how they can be implemented in first step to solving problems is to be able to recog-
GIS. These models are then approached from an nize and describe them. The remainder of the
applications perspective, and three examples are chapter outlines the types and sources of errors in
examined: physical process models; human process GIS to help you identify and deal with problems at
models and decision-making models. To conclude, the appropriate stage of a GIS project. Techniques for
the chapter considers some of the advantages and modelling and managing errors are also considered.
disadvantages of using GIS to construct spatial
process models. Chapter 11: Human and organizational
issues
Chapter 8: Output: from new maps to
enhanced decisions This chapter takes us away from relatively small-
scale research-type applications of GIS, where one
An understanding of the basic principles of map user can be in control from start to finish, and takes
design is essential for the effective communication of a look at some of the issues surrounding larger scale
information and ideas in map form. In addition, an commercial and business applications. In utilities,
understanding of the complexity of the map design local government, commerce and consultancy, GIS
process helps appreciation of the power of maps as must serve the needs of a wide range of users, and
a visualization tool. This chapter considers the should fit seamlessly and effectively into the infor-
advantages and disadvantages of cartographic and mation technology strategy and decision-making
non-cartographic output. In the conclusion to this culture of the organization. In this chapter we
chapter there is a brief discussion of the role of GIS examine the users of GIS and their needs; GIS educa-
output in supporting decision making. tion; how to justify investment in GIS; how to select
and implement a system, and the organizational
I PART 2: ISSUES IN GIS changes that may result.

Chapter 9: The development of computer Chapter 12: GIS project design and
methods for handling of spatial data management

This chapter considers how GIS have developed to How do we understand a problem for which a GIS
their current state. The methods of handling spatial solution is being sought? Two methods are introduced
data that were used before computers were available in this chapter: constructing a rich picture and a root
are examined. These give an insight into what we definition. The method for constructing a GIS data
require computers to do, and how they can help (or model is then discussed. Here a distinction is made
hinder) existing practice. Computer methods for between the conceptual data model and its physical
handling spatial data existed before GIS, so these are implementation in the computer. A closer look at
reviewed, then developments in GIS are discussed the various project management approaches and
together with developments in a selection of com- techniques and tools available for the implementation
plementary disciplines. To conclude we examine of a GIS project follow. Potential implementation
reasons for different rates of growth in different problems and tips for project evaluation are also con-
countries and the role of policy makers in the devel- sidered. To conclude, a checklist is provided to help
opment of GIS. The chapter does not attempt to with the design and implementation of your own GIS

Preface xix

project. A case study is used to illustrate the approach with setting up any GIS project. We hope you enjoy

throughout the chapter. the book!

Chapter 13: The future of GIS Note

GIS of the late 1990s is reviewed here and some of In common with Bonham-Carter (1995) we have
the predictions for GIS in the 1990s are re-visited. found it convenient to use the abbreviation GIS in a
The problems and limitations of current GIS are number of ways:
considered, and from this research and development
issues for the future are identified. The predictions of I to refer to a single system (e.g. a GIS software
other authors and the impact of GIS on our daily package that illustrates this is...)
lives are considered and future scenarios developed.
Finally, the question of whether GIS will still exist in I to refer to several or all geographical information
a few years’ time is addressed. systems (e.g. there are many GIS that...)

I IN SUMMARY I to refer to the field of geographical information
systems (e.g. GIS has a long history), and
We hope that after you have read this book, you will
have the knowledge and enthusiasm to start apply- I to refer to the technology (e.g. the GIS answer
ing GIS in the context of your own course, discipline would be...).
or organization. Whilst the text will not have taught
you how to drive a specific GIS product we hope Trade marks
that it will give you an appreciation of the concepts
on which GIS are based, the methods they use and Mention of commercial products does not consti-
the applications to which they can be put. In addi- tute an endorsement, or, indeed, a refutation of
tion, the book should give you an appreciation of these products. The products are simply referred to
some of the difficulties and considerations associated as illustrations of different approaches or concepts,
and to help the reader relate to the practical world
of GIS.

Ian Heywood, Sarah Cornelius and Steve Carver

Guided tour

Navigation and setting the scene Introduction 171 168 Chapter 5 Data input and editing

3 CHAPTER I INTRODUCTION est, or to identify shortest routes between features CONCLUSIONS
using mobile GIS platforms. As a result they can
Spatial data Chapters 1 to 5 have introduced some of the fun- make decisions about where to go next, and how to This chapter has given an overview of the process of commonly quoted figure suggests that 50–80 per cent
modelling damental concepts behind GIS. The nature of the get there. In a more research oriented application creating an integrated GIS database using the data of GIS project time may be taken up with data encoding
spatial data used has been reviewed, and the encod- ‘what if’ questions might be developed and investi- stream model (Figure 5.1) to guide the process. This and editing. Whilst the amount of time and work
Learning outcomes ing and structuring of these data to produce a gated. For example presentation of the results of has taken us from source data through data encoding, involved is significant it is very important to ensure that
computer representation of the real world out- research into complex planning issues may inform to editing and on to manipulatory operations such as the resulting database is of the best practical quality.
By the end of this chapter you should be able to: lined. We could consider that you now know how a decision-making that will affect the development of re-projection, transformation and rubber sheeting. The quality of the results of any analyses carried out on
I Provide a definition of ‘spatial data model’ car works, how it is built and what fuel to use. Now whole communities and regions. Chapter 1 pre- The chapter has given some insights into how some the GIS will ultimately depend on the quality of the input
I Explain how spatial entities are used to create a data model you are ready to take to the open road and use the sented a list of types of questions that GIS can of these processes are carried out, and highlighted data and the procedures carried out in creating an inte-
I Distinguish between rasters and vectors car to get from one place to another. This chapter address. With analysis we begin to be able to provide issues relevant to data encoding and editing which grated GIS database. This topic will be returned to in
I Describe a spatial data structure begins a journey in GIS, which takes the user from information to help answer some of these. Table 6.1 will be covered again elsewhere in the book. Chapter 10 in a more in-depth discussion of data quality
I Explain what topology is and how is it stored in the computer data to information and ultimately to decision- considers how these questions might be asked in the issues, but first the book continues with details of what
I List the advantages and disadvantages of different spatial data models making. It covers some of the options in GIS for context of two typical GIS applications. Much of the effort required in an individual GIS proj- can be done with the data now integrated in your GIS.
I Discuss the way in which time and the third dimension are handled in GIS data analysis. ect is concerned with data encoding and editing. A
There is a wide range of functions for data analysis
For many users data analysis is the most interest- available in most GIS packages, including measure- REVISION QUESTIONS or longitude, latitude) co-ordinates and entering
ing and rewarding part of a GIS project. This is ment techniques, attribute queries, proximity them into a flat file using keyboard entry. Import
where they can start to find answers to some of their analysis, overlay operations and the analysis of I Describe the main phases of the data stream. the co-ordinates into a GIS (use a spreadsheet
questions, and use GIS to help develop new ques- models of surfaces and networks. To set these in an I How do source data type and nature affect data such as Excel if you don’t have a GIS) and plot
tions for research. The analysis they undertake with applied context, Box 6.1 considers what sort of data them on screen.
GIS may lead to new information that will inform analysis might be required in the imaginary Happy encoding?
decision making. For example, end users of a web- Valley GIS. The questions presented in Box 6.1 will be I Describe the processes of point and stream mode I Choose a stretch of coastline or river and try to
based mapping applications may be able to answer returned to throughout the chapter and suggestions identify the minimum number of points and
simple queries about the location of places of inter- for methods of answering them will be presented. digitizing. their locations that can be used to represent its
I What are the main data sources for GIS? shape. If you can find the same coastline or river
TABLE 6.1 Examples of questions for GIS analysis I What are the main problems associated with using on maps of different scales, repeat this exercise
with these maps and compare different versions
Wind farm siting application Retail store location application the web to find GIS data? to see the effect of scale.
I Why is data editing important?
Location Where is the proposed wind farm location? Where is the region in which the store will be I What methods are available for detecting and FURTHER STUDY – READING
Patterns located?
Trends rectifying errors in: There are a number of GIS texts that cover data
Conditions What is the pattern of land use in this area? What are the flows of traffic along major routes in (a) attribute data, and encoding and editing issues. Hohl (1998) devotes a
this area? (b) spatial data? whole book to the subject of GIS data conversion,
Implications I What are re-projection, transformation and while DeMers (2005) offers two relevant chapters –
How has land use changed over the last 20 years? Which towns in the area have experienced the generalization? on data input and data storage and editing. Jackson
greatest population grown in recent years? I What is geocoding? and Woodsford (1991) present an excellent, if a little
dated, review of hardware and software for data
Where is there an area of privately owned land of Where is there an area of commercially zoned land, FURTHER STUDY – ACTIVITIES capture. Chrisman (1997) offers a reflective review
a suitable size, with adequate access from an close to major transport routes and with potential of techniques and quality issues associated with data
existing road? catchment of over 10,000 customers? I Use the web to try to find GIS data and remotely encoding in a chapter on ‘Representation’. Chapter
sensed imagery for your home town or region. 5 of Martin (1996) and Openshaw (1995) contain
If we located the wind farm here, from where If we locate a store here, what would be the impact overviews of aspects of data encoding issues relevant
would it be visible? on our closest existing store? I Create a flowchart representing the data stream to population census data. For reference to some of
for a GIS project with which you are involved (if the important research work in data encoding you
you are not currently working on a GIS project, need to seek out articles in journals such as the
create a hypothetical example or use one of the
examples in the book).

I Manually digitize a sample set of simple features
(points, lines and areas) from a topographic map
by writing down their easting and northing (x,y

Learning outcomes show Introduction maps out the Conclusions recall and
what you’ll gain from principal themes and content highlight the key issues
reading the chapter for the chapter

Reflection, testing and practice

28 Chapter 1 What is GIS?

and it has been suggested that in certain business dence that many systems fail, and more are under- Reflection box encourages
sectors, innovative flexible organizations with ade- used (Cornelius and Medyckyj-Scott, 1991). So the you to ponder key issues or
quate resources and straightforward applications are issues surrounding how to choose a system and how think further
more likely to succeed (Campbell and Masser, 1995). to implement it successfully require examination.
However, not all GIS are successful. There is evi- These topics are covered in Chapter 11.

REFLECTION BOX

I Look again at Table 1.1. Can you add some Does it match the definitions and component
applications for GIS in areas and activities in descriptions given above? How could it be
which you are interested or involved? improved?
I Try to summarize the components of a GIS in a
I If you have a GIS of your own, or access to one at diagram or table.
work or college, make a list of all its components.

CONCLUSIONS

GIS technology is now well established and, as we will as surveying and field data collection, visualization and
see in Chapter 9, has been in use since the 1960s. database management technology are likely to influ-
Some of the work cited in this and subsequent chap- ence this growth. Further comments on the future of
ters may have been written over 20 years ago – this is GIS can be found in Chapter 13.
an indication of the maturity of GIS. The growth in
application areas and products through the later years There have been some notable failures in GIS.
of the twentieth century has helped GIS to become an Sometimes data difficulties or other technical prob-
accepted tool for the management and analysis of spa- lems have set back system developments and
tial data. This trend is set to continue as computer applications; however, there are also human and
technology continues to improve with faster and more organizational problems at the root of GIS failures.
powerful machines and as more data become available Before we can begin to appreciate these fully, to
in digital formats directly compatible with GIS. In addi- ensure that our GIS applications are successful, it is
tion, the striking advances in related technologies such important to have a good understanding of what a GIS
can do and the data it works with (Chapter 2).

REVISION QUESTIONS I What are the components of GIS?

I What is GIS? Write your own definition of GIS and I Explain the following terms and phrases:
consider why GIS can be difficult to define. 1 Spatial data.
2 Attribute data.
I GIS is routinely used in applications such as siting 3 Spatial referencing.
new industrial developments. What are the 4 Spatial entities.
advantages of using GIS in this context?
I What is the difference between a GIS and
I What type of questions could GIS help geographical information science?
environmental managers address?

End of chapter Revision
Questions help revisit and
test the main themes

Datasets on the website linked to Multiple-choice questions on the website
the Happy Valley Case study provide permit self-testing of basic understanding
an opportunity for practice and and help to build confidence
application using simulated data

Guided tour xxi

Aiding your understanding GIS applications 335 308 Chapter 10 Data quality issues
Theory boxes help explain the thinking
or technology behind a topic, issue BOX 11.1 Corporate, THEORY
or practice multi-department and
independent GIS applications
Visual presentation of material
shows GIS at work in exciting Corporate GIS is developed across an entire organi- Independent GIS exists in a single department and (a) Photograph of Tryfan (b) Ordnance Survey 1:50,000 topographic map
ways and helps demonstrate zation. GIS is usually implemented using a ‘top the GIS will serve the host department’s needs. showing area above 750m
GIS in action down’ approach to promote data sharing, reduce These systems can be adopted quickly and are gen-
data duplication and create a more informed erally task specific. GIS is usually adopted in
decision-making environment. Corporate GIS is response to a specific clearly identified need within
appropriate for a utility company where all depart- the department. In a telecommunications company
ments – customer support, maintenance, research an independent GIS may be used to assist the siting
and development, logistics, sales and marketing – of masts for a mobile phone network.
could benefit from the sharing of data and access to
GIS. Local government agencies also benefit from Corporate and multi-department GIS share many
corporate GIS since all departments work within the benefits, although the benefits will be at a corporate
same geographical area. level where a corporate system is implemented, and
only found in cooperating departments in the multi-
Multi-department GIS involves collaboration department case. These benefits include integration
between different parts of an organization. GIS is of data sets, generation of shared resources, more
implemented in a number of related departments informed decision making, increased data sharing
who recognize the need to share resources such as and improved access to information. Improved con-
data, and the benefits of working together to secure trol over priorities, information and staffing can be
investment for a system. Multi-department GIS may benefits for independent systems.
be appropriate in a retail organization. Marketing,
customer profiling, store location planning, logistics Corporate and multi-department systems can
management and competitor analysis may all make suffer where individual departments have different
use of GIS. Other activities, such as product design priorities, and where there are different levels of
and planning, brand management or financial serv- skills and awareness of GIS and spatial data.
ices, may be completed without the use of GIS. Independent systems may be hampered by lack of
support, isolation and limited lobbying power.

(Source: Hernandez et al., 1999)

purpose and tailored packages for a wide range of successful pioneering GIS applications can provide
application areas. According to GIS World (Anon., high paybacks by giving considerable advantages
1994), the most frequent users of GIS are in govern- over competitors.
ment (over 20 per cent) and education (over 15 per
cent), but there are also users in organizations as Opportunist applications are found in organiza-
diverse as pharmaceuticals, oceanography and tions that kept a careful eye on the pioneers and
restaurants. With such a diverse user group it is not quickly adopted the technology once they saw the
surprising that GIS applications vary enormously in benefits. Following this approach, opportunist
their scale and nature. They can serve the whole organizations let the pioneers bear more of the cost
organization, several departments or just one proj-
ect (Box 11.1). High risk

GIS applications tend to fall into one of three cat- pioneers
egories: pioneering, opportunistic or routine. Pioneering e.g. CGIS
applications are found in organizations that either
are at the cutting edge of their field, or have suffi- opportunists
cient financial reserves to allow them to explore e.g. UK utilities
new opportunities. Pioneering applications are
characterized by the use of unproven methods. As routine users (c) Contour map (d) DEM
such they are considered high-risk and often have e.g. local government
high development costs (Figure 11.1). However, Low risk

Low cost High cost Figure 10.7 Multiple representations of Tryfan, north Wales

Figure 11.1 Development of GIS applications

Applying GIS in the real world

336 Chapter 11 Human and organizational issues Finding and modelling errors in GIS 321 322 Chapter 10 Data quality issues

and risk of application development. Those who business activities from operational, through man- ‘viewpoints’ located in the mixed terrain of BOX 10.7
wait too long risk becoming users of routine GIS agement to strategic (Box 11.2). the Happy Valley ski resort and its surround-
applications. Routine users adopt a tried, tested and ing area. A Monte Carlo simulation is used for example in the snow-covered European Alps. TASK 1: PREPARE TOPOGRAPHIC DATA FOR
refined product with lower risk and cost. In many organizations GIS applications cross the to describe the certainty of the predicted However, when we use low resolution data, the value INPUT TO MONTE CARLO SIMULATIONS
boundaries between the three types of activity intro- radio coverage for each location. Note how of elevation for a single grid cell encompassing a
Whilst there are now plenty of examples of rou- duced in Box 11.2. For example, in a utility uncertainty in the DEM used has the great- large area may be derived in a number of different The first task in preparing topographic data for input
tine and opportunistic GIS applications, truly organization GIS may be used to help operational est effect in areas of relatively flat relief. ways from higher resolution data. Figure 10.19 to Monte Carlo simulations is to characterize the
pioneering applications are rare. This is not surpris- maintenance of pipelines, improve the management shows high resolution data collected for the Zermatt probability distribution which will be applied to the
ing, since few organizations are prepared to be the of marketing services to customers and plan the In both these examples, the results of these region by the Shuttle Radar Topography Mission for low resolution data. For example, regions where ele-
first to apply GIS to new areas because of the high strategic development of the organization. In many simulations can be overlaid on the normal, the Swiss Alps, and a corresponding generalization vation varies little will have low standard deviations in
risks and costs that may be involved. For example, in organizations a GIS will initially provide information unvaried results as a visual means of commu- of these terrain data to 20 km (a resolution typically elevation – for example the Prairies of Mid-West
our nuclear waste disposal example, NIREX was in support of one activity and, if it is successful, nicating the likely effects of input data error. used in an ice sheet model). America. Mountainous regions such as the Alps have
unlikely to risk any adverse reaction or publicity to develop so that information can be provided to sup- The main drawback with this method is the high standard deviations in elevation.
such a sensitive problem by application of what was, port other activities. The evolution of CGIS, the large number of simulations required to This case study illustrates the use of Monte Carlo
in the mid-1980s, a new technology. Despite being, by Canadian GIS, illustrates this development. achieve a statistically valid result. For GIS simulation to explore the dependence of model Resampling is a standard technique in GIS to
today’s standards, a relatively straightforward GIS processes where the effects of error need to results on uncertainties in elevation values. Monte derive rasters of lower resolutions. For example, to
application, the use of GIS to perform site search and CGIS underwent three major stages of develop- be estimated, up to 100 extra sets of process- Carlo simulations require that a model is run many resample from a resolution of 100 m to 1 km, the
evaluation analysis was, at the time, the realm of aca- ment (Crain, 1985). The system was used first in an ing may be required. Monte Carlo simulation times with one variable being varied according to origin of the new raster must first be selected. Then,
demic research (Openshaw et al., 1989; Carver, 1991b). operational role for data inventory, with data collec- is only likely to be used for very important some probability distribution, in order to explore the values for the new raster at a spacing of 1 km must be
tion and querying the dominant functions. Then analyses requiring a high level of confidence model’s sensitivity to that variable. In the following identified. Typically, a neighbourhood function is
Pioneering, opportunistic and routine applica- CGIS became more of a management tool with new in the results. Although initial research on example, the use of standard GIS techniques to pre- applied to the source resolution raster to derive a
tions of GIS have permeated the entire range of data being produced through the use of spatial analy- Monte Carlo simulation focused on its use pare multiple surfaces representing uncertainty value at the new target low resolution. The neighbour-
for modelling positional error in vector map in elevation is explained. The approach adopted hood function may, for example, take the mean of all
overlay (Openshaw et al., 1991), the approach involved four tasks. 100 cells inside the 1 km cell, or may apply some
is equally applicable to modelling attribute or neighbourhood function to a subset of these cells.
positional error in raster or vector GIS opera-
BOX 11.2 Business PRACTICE tions. Another example of the use of Monte
applications of GIS CASE Carlo simulation modelling to prepare mul-
STUDY tiple surfaces representing uncertainty in
OPERATIONAL APPLICATIONS I an education department might use GIS to pro- Figure 10.18 Simulating the effects of DEM error and algorithm terrain data is provided in Box 10.7.
duce options for school closures when deciding uncertainty on radio communications in the Happy Valley area
Operational GIS applications are concerned with how to distribute limited resources; or
managing facilities and assets. For example: BOX 10.7 Uncertainty and
I a telecommunications company might use GIS to terrain data in ice-sheet
I a utility company may use GIS to identify assets identify and evaluate possible communications mast modelling
such as pylons and masts in need of routine main- sites to serve the maximum possible population.
tenance; Ross Purves and Felix Hebeler Any abstraction cannot completely represent a (a) (b)
STRATEGIC APPLICATIONS system and thus differences between a modelled
I a waste management company may use GIS to Strategic GIS applications are concerned with the Ice sheet modelling seeks to represent the behaviour scenario and ‘reality’ must exist. It is important to Figure 10.19 Resampling SRTM data from 100 m to 20 km resolution. (a) A view of Zermatt from the east at 100 m
route waste collection vehicles; or creation and implementation of an organization’s of ice sheets under past, present or future climates. understand these differences, or uncertainties, in resolution (top) and 20 km resolution (bottom). (b) DEM of the European Alps (top) and graph of standard deviation
strategic business plan. For example: It is often based on some form of proxy record, for modelled results to make use of models as tools in elevation with respect to elevation for resampled data (bottom)
I a property management company may maintain example the temperature signal extracted from an ice for understanding and predicting the behaviour of
hotel buildings and grounds with the help of maps I a ski equipment retailer may decide which geo- core record. The data obtained can help us explore natural systems. One source of uncertainty arises
and plans from its GIS. graphical areas to target over the next five years concerns about future climate change such as the because ice sheet models are typically run at low
after geodemographic analysis using GIS; impact of climate on ice extents and thus potential resolutions, so certain processes are not well
MANAGEMENT/TACTICAL APPLICATIONS changes in sea level and the influence of melt water resolved. For example, inception (the formation of
I a local government organization may decide on inputs to ocean systems. new ice sheets) can only take place in regions where
Management (or tactical) GIS applications are con- budget reallocations after analysis of population the annual mass balance is positive (that is more
cerned with distributing resources to gain competitive growth or decline in certain areas using modelling Ice sheet modelling, in common with all numeri- snow accumulates over a year than melts).
advantage. For example: and GIS; or cal modelling, seeks to improve our understanding Such regions are typically at high elevations –  
of the real world through an abstraction of reality.
I a ski holiday company may use GIS to identify I a catering business may decide to expand a
appropriate potential customers to receive direct restaurant chain to 100 outlets after analysis of the
mailings; location of its competitors with GIS.

Practice boxes give detailed Multi-disciplinary case studies show how GIS principles and
examples of GIS being used in technology are used in multiple ways, by different people, and
the real world across a range of industries

Further Study

106 Chapter 3 Spatial data modelling End of chapter activities and 132 Chapter 4 Database management
revision questions for self-study
REVISION QUESTIONS FURTHER STUDY – READING or discussion encourage the WEB LINKS
application and practice of knowledge
I Describe how the raster and vector approaches Peuquet (1990) provides an excellent description of and research beyond the textbook Definitions and concepts: Online databases:
are used to construct point, line and area entities the stages involved in moving from a conceptual
for representation in the computer. spatial model of the real world through to its repre- I Data models http://unixspace.com/ Source OECD
sentation in the computer. Burrough (1986) and context/ databases.html http://caliban.sourceoecd.org/
I What are data structures? Outline their Aronoff (1991) provide good introductions to data vl=2215771/cl=97/nw=1/rpsv/home.htm
importance in GIS. modelling, particularly with regard to the distinc- Database design:
tion between the raster and vector data models. Transport tracking:
I What is a TIN? How are TINs constructed? Wise (2002) offers more detail on raster, vector, sur- I Unesco training materials on data modelling
face and network models for GIS. Burrough (1986) http://ioc.unesco.org/oceanteacher/ I Nextbus
I What are surface significant points and how can also provides a good introduction to the TIN data resourcekit/Module2/GIS/Module/ http://www.nextbus.com/
they be defined? model and extends the discussion to look in more Module_e/module_e2.html predictor/newUserWelcome.shtml
detail at digital elevation models.
I How can networks be modelled in GIS? I New York State Archive, GIS development I Surrey, UK
Laurini and Thompson (1992) take a look at the guidelines http://www.acislive.com/pages/
I Why is it difficult to model the third and fourth spatial data modelling process from the viewpoint of http://www.nysarchives.org/ a/ busnet. asp?SysID=13
dimensions in GIS? a computer scientist rather than a geographer. They nysaservices/ns_mgr_active_gisguides_
provide a good introduction to the concept of dbplanning.shtml Visit our website http://www.pearsoned.co.uk/heywood
FURTHER STUDY – ACTIVITIES object-oriented database design for handling spatial for further web links, together with self-assessment
information. However, material about data model- questions, activities, data sets for practice opportunities
I Draw a simple polygon on a coarse grid (approxi- ling is dispersed throughout their book, so it may and other interactive resources.
mately 10 × 10 cells) and use different raster data take you some time to find what you are looking for.
encoding methods to represent this as a text file. If you want to dig deeper into spatial data structures Reading and Web Links help you
then Worboys and Duckham (2004) gives a clear and investigate key topics of further
I Draw a more complex polygon on a piece of readable account of a range of different approaches. interest
graph paper and encode this using the area or Samet (1989) provides more advanced material, par-
region quadtree method. Represent this as tree ticularly on quadtree structures.
data structure on another piece of paper.
To explore the history and issues associated with
I Make a TIN model from a sheet of paper (see Box the evolution of 3D GIS the paper by Raper and Kelk
3.4). (1991) is a good introduction. Langran (1992) pro-
vides a comparable introduction to the issues
I ‘Digitize’ a simple polygon network using surrounding the representation of time in the GIS
spaghetti and point dictionary methods. environment.

I Select a small area of contours and spot heights Aronoff S (1991) Geographic Information Systems: A
from a topographic map. Try creating a TIN from Management Perspective. WDL Publications, Ottawa
these data using one of the methods described in
Box 3.6. Burrough P A (1986) Principles of Geographical
Information Systems for Land Resources Assessment.
I Use free surface modelling software (e.g. LandSerf Clarendon Press, Oxford
– see web link below) to load and explore some
surface data. Langran G (1992) Time in Geographical Information
Systems. Taylor and Francis, London

Case studies Location Sector

Case study

Any
Czech Republic
UK
Europe
Ghana, Africa
The Netherlands
India
Remote areas
Australia
Canada, N. America
London
USA
Tourism
Environmental planning
Commercial planning
Transport planning
Planning
Social/public sector
Business
Hazard assessment
Local government
Environmental Mgmt.
Urban planning

1.1 Questions GIS could answer in Happy Valley o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

1.2 Issues raised by the NIREX case study o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

1.3 Issues raised by the Zdarske Vrchy case study o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

1.4 Issues raised by the house hunting case study

Steve Carver, University of Leeds and o
Sarah Cornelius, University of Aberdeen

2.6 Mapping the UK’s railway o
Jackie Brennan, Network Rail

2.8 High resolution data solutions to complex

modelling problems
Stuart Lane, Durham University and Joe Holden, o
Leeds University

3.8 Monitoring coastal change with GIS o
Jonathan Raper, City University, London

4.4 Encroachments on public land in Accra, Ghana o
Isaac Karikari, Lands Commission, Accra, Ghana

4.6 Delivering census interaction data: WICID oo
John Stillwell, University of Leeds

5.2 Business process outsourcing and the Indian o
GIS industry o
Ankur Das, India oo

5.4 Fireguard: using GIS and remote sensing for
fire management
Jim Hogg, University of Leeds

5.8 Going Dutch with small scale topography
Martin Gregory, Laser-Scan

5.9 Ordnance Survey (OS) data collection – from oo
plan to MasterMap
Paul Corcoran, Ordnance Survey

5.10 An integrated GIS for Happy Valley o
Steve Carver, University of Leeds and o
Sarah Cornelius, University of Aberdeen
o
6.1 Data analysis in Happy Valley
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

6.3 Modelling remoteness in mountain areas
Steve Carver and Steffen Fritz, University of
Leeds

6.5 Siting a nuclear waste repository o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

6.6 Raster analysis of the nuclear waste siting o
example
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

Case studies xxiii

Case study Location Sector

Any
Czech Republic
UK
Europe
Ghana, Africa
The Netherlands
India
Remote areas
Australia
Canada, N. America
London
USA
Tourism
Environmental
Environmental planning
Commercial planning
Transport planning
Planning
Social/public sector
Business
Hazard assessment
Local government
Environmental Mgmt.
Urban planning

6.9 Identifying mountains with GIS o
Jo Wood, City University, London o

6.10 Wind farm visibility in Cape Bridgewater,
Australia
Ian Bishop, University of Melbourne

7.1 The development of a snow cover model o
for Happy Valley
Steve Carver, University of Leeds and oo
Sarah Cornelius, University of Aberdeen oo

7.3 Using GIS to help manage air quality o
through estimating emissions of o
atmospheric pollutants
Sarah Lindley, University of Manchester o

7.5 GIS-enabled retail sales forecasting o o
Tony Hernandez, Ryerson University
o
7.8 Fuzzy GIS and the fear of crime
Andrew Evans, University of Leeds

8.2 Retail geovisualization: 3D and animation o
for retail decision suppor o
Tony Hernandez, Ryerson University

8.4 Virtual London
Michael Batty, University College London

8.5 Happy Valley Internet GIS o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

8.7 Public participation GIS
Richard Kingston, University of Manchester

9.2 Developments in computer cartography o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen o

9.3 CGIS: an early GIS
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

9.5 CORINE: an international multi-agency
GIS project
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

9.6 GIS and North West Water o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

10.7 Uncertainty and terrain data in
ice-sheet modelling
Ross Purves and Felix Hebeler, University
of Zurich

10.9 Errors in the nuclear waste case study o
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

11.6 Siting a nuclear waste repository
Steve Carver, University of Leeds and
Sarah Cornelius, University of Aberdeen

Acknowledgements

During the production of three editions of this book We would like to thank the reviewers who have
many, many people have helped with support, provided valuable feedback, advice and suggestions
advice and ideas both directly and indirectly. We have for this new edition. Our thanks go to:
taken the opportunity here to thank some of those
who have helped, but we recognize that we have Darius Bartlett, University College Cork, Ireland
omitted many others who deserve appreciation.
Josie Wilson, Sheffield Hallam University
For help with the text of the first edition we
should thank Steve Wise and an anonymous Dr Robert MacFarlane, Northumbria University,
reviewer who made constructive comments on early UK
drafts. Tony Hernandez, Chris Higgins, Derek Reeve,
David Medyckyj-Scott and Tony Heywood are also Dr Duncan Whyatt, Lancaster University, UK
thanked for reading drafts of various sections. Help
with the diagrams, plates and cartoons and word Steve Wise, University of Sheffield, UK
processing came from Gustav Dobrinski, Bruce
Carlisle, Simon Kenton and Tracey McKenna. Thomas Balstroem, University of Copenhagen,
Denmark.
All of the case studies used in the book are the
result of the work of teams of researchers and aca- Anne Lucas, Geography Department, University
demics. Stan Openshaw and Martin Charlton were of Bergen
involved with the original radioactive waste siting
study. Jim Petch, Eva Pauknerova, Ian Downey, Mike Angela Cuthbert, Millersville University, USA
Beswick and Christine Warr worked on the Zdarske
Vrchy study. The initial ideas for the House Hunting Tim Elkin PhD, Camosun College, Victoria, BC,
game came from a team that included James Oliver Canada
and Steve Tomlinson. This last case study was made
available by the GeographyCAL initiative based at Christine Earl, Carleton University, Canada
the University of Leicester following additional
input and support from Roy Alexander, John Dr Jane Law, University of Cambridge, UK
McKewan and John Castleford. In particular, we
would like to thank the many people who have Dr Adrian Luckman, Swansea University, UK
given their time freely to write the excellent new
case studies you see in this third edition. These Dr H.A. Oosterhoff, University of Groningen,
people who friends, colleagues, associates and stu- The Netherlands
dents who have worked with us on GIS projects and
have allowed us to pass on the richness and variety Bernd Etzelmuller, University of Oslo, Norway.
of GIS applications across the globe. They are all
introduced in the author bibliographies. At Addison Wesley Longman, Prentice Hall and
Pearson Education we have worked with many
Many other individuals have supported our work individuals. Particularly we should thank Vanessa
on this book – Helen Carver, Thomas Blaschke and Lawrence for help and advice at the start of the
Manuela Bruckler deserve special mention. In addi- project, Sally Wilkinson, who took over during the
tion, the Universities of Leeds, Aberdeen and Salzburg, middle stages, and Tina Cadle-Bowman, Matthew
the Manchester Metropolitan University and the Vrije Smith, Shuet-Kei Cheung and Patrick Bonham,
Universiteit Amsterdam have allowed us to use their who steered the book towards publication. For
resources to help compile the materials. their work on the second edition thanks are due to
Matthew Smith, Bridget Allen, Morten Funglevand,
Paula Parish and Nicola Chilvers. For their support
and innovative ideas during the preparation of this
third edition we must thank Andrew Taylor, Janey
Webb, Sarah Busby, Maggie Wells and Joe Vella. In
addition we have been supported by a team of pic-
ture researchers and designers who have helped to
create the striking new product you see.

Acknowledgements xxv

And lastly, we should thank all our students and ular mention – Jim Petch, Derek Reeve, Josef Strobl,
colleagues. We started this project to help our stu- Nigel Trodd, Christine Warr, Steve Tomlinson and
dents understand the mysteries of GIS. Some have Bev Heyworth. We apologize to them unreservedly
been an inspiration, and gone on to exciting and if they see ideas and examples inspired by their
rewarding careers in the GIS field. Others struggled course materials within the book!
with basic concepts and ideas, and helped us to real-
ize just what we needed to take time to explain. We hope that the experience we have gained
Writing for distance learning students has taught us between the initial idea for the book and publication
all a tremendous amount and our colleagues from of the third edition has helped to make this a useful
the UNIGIS distance learning course deserve partic- and readable introduction to GIS.

Ian Heywood, Sarah Cornelius and Steve Carver

Publisher’s acknowledgements

We are grateful to the following for permission to HMSO. © Crown Copyright 2006. All rights
reproduce copyright material: reserved. Ordnance Survey Licence number
100030901; Figure 2.19a: Courtesy of the United
Figure 1.2: Richard Armstrong; Figure 1.3b: Getty States Department of Agriculture; Figure 2.19b:
Images; Figure 1.3c,d: British Nuclear Fuels Ltd; Copyright © Bluesky, used by permission; Figure
Figure 1.7a: British Geological Survey (www.bgs. 2.19d: Courtesy of the United States Geological
ac.uk). Copyright © NERC, IPR/71-39C, reproduced Survey (USGS); 2.21a: CNES 1994-Distribution SPOT
by permission of the British Geological Survey; Image; Figure 2.21b: i-cubed: LLC; Figure 2.21c:
Figure 1.7b: From the National Statistics website Science Photo Library/Earth Satellite Corporation;
(www.statistics.co.uk). Crown copyright material is Figure 2.21d: Wetterzentrale, Germany; Figure 2.21e:
reproduced with the permission of the Controller of i-cubed: LLC; Figure 2.22a-c: PTS (Precision Terrain
HMSO; Figure 1.7d: Joint Nature Conservation Surveys); Figure 2.26a: Alamy/Widstock; Figure
Committee (www.jncc.gov.uk); Figure 1.14a: Courtesy 2.26b: Empics; Figure 2.27b,c: Magellan; Figure 3.5a–c:
of Integraph; Figure 1.14c: Alamy/ ScotStock; Figure From The Digital Chart of the World. Courtesy of ESRI.
1.14d: Courtesy of Main StreetGIS; Figure 1.14e: Copyright © ESRI. All rights reserved; Figure 3.17a:
Courtesty of Trimble; Figure 1.15a,b: Screenshot Reproduced by permission of Ordnance Survey on
shows ESRI Graphical User Interface (GUI). behalf of HMSO. © Crown Copyright 2006. All
ArcMap/ ArcView/ArcInfo Graphical User Interface rights reserved. Ordnance Survey Licence number
is the intellectual property of ESRI and is used 100030901; Figure 3.20: Courtesy of NASA/JPL/NIMA;
herein with permission. Copyright © 2005 ESRI all Figure 3.21a,b: Courtesy of NASA/JPL/NIMA; Figure
rights reserved; Figure 1.19: Courtesy of the United 3.21c: Environment Agency Science Group; Figure
States Geological Survey (USGS); Figure 1.20: 3.23: Courtesy of the United States Geological
Courtesy of the United States Geological Survey Survey (USGS); Figure 3.27a,b: Courtesy of the
(USGS); Figure 2.2a: Published by the Met Office, United States Geological Survey (USGS); Figure
top: © Crown copyright 2006, bottom: © copyright 3.29a: London Underground map from Transport for
EUMETSAT /Met Office 2006; Figure 2.2b: Alamy/ London, with permission of Transport Trading
Dominic Burke; Figure 2.4a: Maps reproduced by Limited; Figure 3.29b: From www.simonclarke.org,
permission of Ordnance Survey on behalf of HMSO. © Simon Clarke 2000; Figure 3.33a: Rockware, Inc.,
© Crown Copyright 2006. All rights reserved. with permission; Figure 3.33b: Centre for Advanced
Ordnance Survey Licence number 100030901 and Spatial Analysis (CASA) University College London;
image shown on map cover courtesy of Peter Baker/ with permission; Figure 4.1a: Screenshot shows ESRI
International Photobank; Figure 2.5: Topfoto/Roger- Graphical User Interface (GUI) ArcMap/ArcView/
Viollet; Figure 2.10a–g: Reproduced by permission of ArcInfo Graphical User Interface is the intellectual
Ordnance Survey on behalf of HMSO. © Crown property of ESRI and is used herein with permis-
Copyright 2006. All rights reserved. Ordnance Survey sion. Copyright © 2005 ESRI all rights reserved;
Licence number 100030901; Figure 2.12a: Courtesy of Figure 4.1b: Screenshot shows IDRISI software
NASA; Figure 2.12b,d,e: From John Savard’s home- interface. Reproduced with permission from Clark
page (www.members. shaw.ca/quadibloc/maps/mcy Labs, Clark University; Figures 4.10 and 4.11: Data
0103.htm) by permission of John Savard; Figure from Karikari, I., Stillwell, J. and Caver, S. (2005)
2.12c: From the Illinois State University, microcam The application of GIS in the lands sector of a
website, (www.ilstu.edu/ microcam/map_projections/ developing country: challenges facing land admin-
Conic/Lambert_Conformal_Conic.pdf) by permis- istrators in Ghana, International Journal of Geographical
sion of Dr Paul S. Anderson; Figure 2.17a Reproduced Information Science, Vol. 9 No. 3, Taylor & Francis
by permission of Ordnance Survey on behalf of Ltd, with permission (www.tandf.co.uk/journals);

Publisher’s acknowledgements xxvii

Figure 4.16: From the 2001 Census: Special Workplace 100030901; Figure 10.9: Screenshot shows IDRISI
Statistics (Level 1), National Statistics website interface. Reproduced with permission from Clark
(http://statistics. co.uk). Crown copyright material is Labs, Clark University; Figure 13.1a–c: Courtesy of
reproduced with the permission of the Controller of Encom Technology Pty Ltd; Figure 13.3: Gordon Sell
HMSO; Figure 5.3: From Able Software Corporation, Public Relations (www.gspr.com), with permission;
(www.ablesw. comr2v/r2v3.jpg), used by permission; Figure 13.6: From the Houston Transtar website,
Figure 5.6: Screenshot shows ESRI Graphical User (http://traffic.houstontranstar.org/layers/layers.
Interface (GUI) ArcMap/ArcView/ArcInfo Graphical aspx), with permission of the Houston Transtar
User Interface is the intellectual property of ESRI Consortium; Figure 13.7: From the ESRI website
and is used herein with permission. Copyright © (www.esri.com/news/pressroom/graphics/Katrina/
2005 ESRI all rights reserved; Figure 5.7a: CST flooding0831estimate_lg.jpg). Image Copyright ©
Germany, used by permission; Figure 5.7b: Epson DigitalGlobe, used with permission. The ArcMap/
(UK) Ltd, used by permission; Figure 5.08: Courtesy ArcView/ArcInfo Graphical User Interface is the
of the United States Geological Survey (USGS); intellectual property of ESRI and is used herein with
Figures 5.12 and 5.13: Laser-Scan. Copyright © 2005 permission. Copyright © ESRI; Figure 13.8a: From
LS 2003 Ltd. All rights reserved; Figure 5.15: Courtesy the ESRI website (www.esri.com/news/pressroom/
of Peter H. Dana; Figure 5.16a–f: From The Digital graphics/Katrina3_lg.jpg). Image Copyright ©
Chart of the World. Courtesy of ESRI. Copyright © 2005– 2006 Data Transmission Network Corporation,
ESRI. All rights reserved; Figure 5.16 inset: From d/b/a Meteorlogix. The ArcMap/ArcView/ArcInfo
ESRI, ArcGIS online help system, courtesy of ESRI. Graphical User Interface is the intellectual property
Copyright © 2005 ESRI. All rights reserved; Figure of ESRI and is used herein with permission.
5.20 and 5.21: Reproduced by permission of Copyright © ESRI; Figure 13.8b: Courtesy of NASA;
Ordnance Survey on behalf of HMSO. © Crown Figure 13.8c: From the Hurrevac website, www.hurre-
Copyright 2006. All rights reserved. Ordnance vac.com. used by permission of Sea Island Software;
Survey Licence number 100030901; Figures 6.11 and Figure 13.9a,b: DigitalGlobe (www.digitalglobe.com/
6.12a–c: Courtesy of the United States Geological katrina_gallery.html) with permission; Figure 13.10:
Survey (USGS); Figure 6.24: Horizon Simulation Reproduced from http:// arcweb.esri. com/sc/hurri-
Ltd, with permission; Figure 6.27a,b: Institute of cane_viewer/index.html, Copyright © 2005 ESRI,
Cartography, ETH Zurich, Switzerland, used by per- TANA. Used with permission; Figure 13.11a,b:
mission; Figures 6.30 and 6.31: Courtesy of Ordnance DigitalGlobe (www.digitalglobe.com/tsunami_gallery.
Survey (OS); Figure 7.15: Screenshot shows IDRISI html), with permission; Figure 13.13a: From the Earth
interface. Reproduced with permission from Clark Observatory/NASA website. http://earthobservatory.
Labs, Clark University; Figures 8.8 and 8.9: nasa.gov; Figure 13.13b: From the Natural
Screenshots show ESRI Graphical User Interface Environment Research Council website, www.nerc.
(GUI) ArcMap/ArcView/ArcInfo Graphical User ac.uk/publications/latestpressrelease/2000-13-1icemelt.
Interface is the intellectual property of ESRI and is asp, copyright © Jonathan L Bamber, used by permis-
used herein with permission. Copyright © 2005 ESRI sion; Figure 13.14a: Earch Observation Research and
all rights reserved; Figure 8.10a,b: From Virtual Field Application Center (EORC) website, www.eorc.
Course website (www. geogle.ac.uk/vfc/). We wish to jaxa.jp/en; Figure 13.14b: MECAA project website,
acknowledge the Department of Geography, www.unites.uqam.ca/ mecaa; Figure 13.14c from
University of Leicester for its permission to use this LANDSAT Pathfinder satellite images.
material; Figure 10.6: Martin Sauerbier, Institute of
Geodesy & Photogrammery; Figure 10.7b: Reproduced In some instances we have been unable to trace
by permission of Ordnance Survey on behalf of the owners of copyright material, and we would
HMSO. © Crown Copyright 2006. All rights appreciate any information that would enable us
reserved. Ordnance Survey Licence number to do so.

Abbreviations and acronyms

1:1 One-to-one relationship EDM Electronic Distance Metering
1:M One-to-many relationship ESMI European Spatial Meta-information
2.5D Two-and-a-half-dimensional system
2D Two-dimensional ESRC Economic and Social Research
3D Three-dimensional Council (UK)
4D Four-dimensional ESRI Environmental Systems Research
AGI Association for Geographical Institute
Information (UK) ETHICS Effective Technical and Human
AI Artificial Intelligence Implementation of Computer-based
AM/FM Automated Mapping/Facilities EUROGI Systems
Management European Umbrella Organization for
BGS British Geological Survey GAM Geographical Information
BS British Standard GBF-DIME Geographical Analysis Machine
BSU Basic Spatial Unit Geographical Base File, Dual
CAD Computer-aided Design GDF Independent Map Encoding
CAL Computer-aided Learning GeoTIFF Geographic Data File
CASE Computer-assisted Software GIS Geographic Tagged Image Format File
Engineering GISc Geographical Information System
CCTV Closed-circuit Television GLONASS Geographic Information Science
CD-ROM Compact Disc – Read Only Memory Global Orbiting Navigation Satellite
CEN Comité Européen de Normalisation GML System
(European Standards Commission) GPS Geography Markup Language
CGIS Canadian Geographic Information GUI Global Positioning Systems
System HTTP Graphical User Interface
CORINE Coordinated Information on the IAEA Hypertext Transfer Protocol
European Environment International Atomic Energy
CoRWM Committee on Radioactive Waste ID Authority
Management IS Identifier
CRT Cathode Ray Tube ISO Information System
CUS Census Use Study (USA) IT International Standards Organization
DBMS Database Management System LAN Information Technology
DEM Digital Elevation Model LBS Local Area Network
DIGEST Digital Geographic Information LCD Location-based Services
Exchange Standards LiDAR Liquid Crystal Display
DNF Digital National Framework M:N Light Detection and Ranging
DoE Department of the Environment (UK) MAUP Many-to-many relationship
DOS Disk Operating System MCE Modifiable Areal Unit Problem
dpi Dots Per Inch MEGRIN Multi-criteria Evaluation
DSS Decision Support System Multipurpose European Ground
DTM Digital Terrain Model MSS Related Information Network
DXF Data Exchange Format NAVSTAR Multispectral Scanner
EAM Entity Attribute Modelling Navigation System with Time and
ED Enumeration District (UK) NCDCDS Ranging
EDA Exploratory Data Analysis National Commitee on Digital
Cartographic Data Standards (USA)

NCGIA National Center for Geographic SDTS Abbreviations and acronyms xxix
Information and Analysis (USA) SLC
NERC Natural Environmental Research SPOT Spatial Data Transfer Standard
Council (UK) SQL System Life Cycle
NII Nuclear Installations Inspectorate (UK) Système pour l’Observation de la Terre
NIMBY Not In My Back Yard SSA Standard Query Language (or
NIREX Nuclear Industry Radioactive Waste SSADM Structured Query Language)
Executive (UK) Soft Systems Analysis
NGDC National Geospatial Data SWOT Structured Systems Analysis and
Clearinghouse Design
NJUG National Joint Utilities Group (UK) TCP/IP Strengths, Weaknesses, Opportunities,
NTF National Transfer Format (UK) Threats
NTF Neutral Transfer Format (UK) TIGER Transmission Control
NWW North West Water plc Protocol/Internet Protocol
OCR Optical Character Recognition TIN Topologically Integrated Geographic
OGC Open GIS Consortium TM Encoding Reference File (USA)
OGIS Open Geodata Interoperability TV Triangulated Irregular Network
Standards URL Thematic Mapper
OO Object-oriented UK Television
OS Ordnance Survey (UK) URISA Uniform Resource Locator
PC Personal Computer United Kingdom
PDA Personal Data Assistant USA Urban and Regional Information
PERT Program Evaluation and Review UTM Systems Association (USA)
Technique VIP United States of America
PHP Hypertext Preprocessor VR Univeral Transverse Mercator
PLSS Public Land Survey System (USA) VRML Very Important Point
PPGIS Public Participation GIS WALTER Virtual Reality
REGIS Regional Geographic Information WAP Virtual Reality Modelling Language
Systems Project (Australia) WAAS Terrestrial Database for Wales
RINEX Receiver Independent Exchange Format WWW Wireless Application Protocol
RRL Regional Research Laboratory (UK) XML Wide Area Augmentation System
SDSS Spatial Decision Support System ZVI World Wide Web
Extensible Markup Language
Zone of Visual Influence

About the authors Ian Bishop is a Professor of Geomatics
at the University of Melbourne. His
Book authors research areas include environmental
visualization, landscape assessment
Ian Heywood is Director of Skills and Learning and decision support tools for land use
for Scottish Enterprise Grampian. He previously change. He is currently working with
worked in private consultancy and as Director of Geographic Information Systems (GIS), visualization
Open and Distance Learning for Robert Gordon technology and agent-based models to improve the
University. He has undertaken research in GIS management of natural resources and to contribute
and taught at Universities in the UK, Canada and to public participation in planning and design issues.
Netherlands. Ian has a PhD from the University of (See Box 6.10: Wind farm visibility in Cape
Newcastle upon Tyne and is a Visiting Professor at Bridgewater, Austrialia, p. 211.)
Abertay University, Dundee.
Jackie Brennan is Spatial Data
Sarah Cornelius is a Lecturer in the School of Services Manager with Network Rail.
Education at the University of Aberdeen and an She is responsible for managing the
Associate Lecturer for the Open University where company’s digital mapping agree-
she specializes in adult and online learning. She pre- ment with Ordnance Survey and
viously taught GIS at Manchester Metropolitan managing data conversion and main-
University and the Free University of Amsterdam, tenance teams for the corporate GIS. Jackie holds a
and has also worked as an independent consultant Master’s degree in Geography from University
developing education and training resources for pri- College Dublin and a post-graduate diploma in busi-
vate sector organizations developing and using GIS. ness management. After university she worked in
Dublin as a market research data analyst. In 1999 she
Steve Carver is a Senior Lecturer in the Department moved to London and took up a position in GI data
of Geography at the University of Leeds. He is management with a GIS consultancy company based
Director of the MA in GIS and MSc in GIS. Steve is in central London. In 2002, she joined Network Rail
an active researcher and has undertaken field based and began work in the company’s property sector.
GIS projects in Siberia, Greenland and the European (See Box 2.6: Mapping the UK’s railway, p. 53.)
Alps. Steve has a PhD in nuclear waste siting.
Paul Corcoran joined the Ordnance
Case study authors Survey, Great Britain’s National
Mapping Agency, in 1984 and has
Michael Batty is Bartlett Professor undertaken various surveying and
of Planning at University College management roles within the
London where he directs the Centre Agency. Paul has predominantly
for Advanced Spatial Analysis worked in the North of England and is currently
(CASA). His research is focused on based in the Leeds office. He was sponsored by the
new approaches to urban simulation OS to undertake a BA (Hons) degree in Business
modelling, with a strong focus on visualization. He
is the Editor of Environment and Planning B. (See Box 8.4:
Virtual London, p. 269.)

About the authors xxxi

Management, at Leeds Metropolitan University in Since joining Laser-Scan in 1989
1997, graduating in 2001. After 18 months acting as a Martin Gregory has worked in a wide
Technical Advisor, he returned to academia and variety of roles. Initially developing
completed an MSc in GIS for Business and Service a solid understanding of digital carto-
Planning at the University of Leeds in August 2005. graphy and capture systems he
(Box 5.9: Ordnance Survey (OS) data collection – progressed into technical support, and
from Plan to MasterMap, p. 162.) then post sales support and consultancy roles. More
recently he spent 2 years based at the Mexican
Ankur Das is a Geography graduate National Mapping Agency as a project manager,
with a strong interest in environmen- responsible for the maintenance of multi-vendor soft-
tal issues confirmed by the choice of a ware and hardware used in the geospatial data
postgraduate degree in Geographical production flowlines. Upon returning to the UK,
Information Systems at the University Martin got involved with Laser-Scan’s Oracle based
of Leeds in the UK. He is presently Radius Topology product and is now a Technical Sales
working professionally in India. His research interests Consultant, working primarily in the Intergraph
include: enhanced visualization and spatial algo- partner channel to develop and support sales of
rithms, image processing of satellite remote sensing Radius Topology worldwide. (See Box 5.8: Going
data, change detection from satellite imagery, retail Dutch with small scale topography. p. 153.)
decision support mechanism from the perspectives of
a GIS, sustainable development and economic growth, Felix Hebeler studied Geography
historical GIS. (See Box 5.2: Business process outsourc- and Biology (at the University of
ing and the Indian GIS Industry, p. 140.) Giessen, Germany and Rhodes
University, RSA) with an emphasis
Andrew Evans is a lecturer in on environmental and ecological
the School of Geography at the issues and a focus on modeling using
University of Leeds, where he teaches Geographical Information Systems. He is currently
Masters courses in GIS, AI and com- studying for a PhD at the University of Zurich with
puter programming. His interests his research interests including environmental and
include building systems to enable ice sheet modeling as well as analysis and generalisa-
online democracy and modelling social interactions, tion of digital elevation models. His ongoing PhD
though he has sidelines in glaciology and Dark Age project is investigating the influence of topographic
mythopoetic landscapes. (See Box 7.8: Fuzzy GIS and representation on uncertainties in large scale envi-
the fear of crime, p. 243.) ronmental model results. (See Box 10.7: Uncertainty
and terrain data in ice-sheet modelling, p. 321.)
Steffen Fritz studied Physics and
Geography at the University of Dr Tony Hernandez is the Director
Tübingen and got his Master of of the Centre for Study of Commercial
Science from the University of Activity and Associate Professor in
Durham in 1996. He undertook his the Department of Geography at
PhD in the field of wild land research Ryerson University, Toronto. Tony
at Leeds University. Since 2001 he has worked at the teaches courses at the undergraduate
Joint Research Centre of the European Commission and graduate level in the area of business geomatics,
in the field of global land cover mapping, monitor- with focus on the retail and service economy. His
ing agriculture and food security. In the summer of research interests include geographic information
2004 he worked on the production of a global bio- system, spatial analysis, geovisualization and deci-
mass map at the International Institute of Applied sion support. (See Box 7.5: GIS-enabled retail sales
Systems Analysis in Vienna. (See Box 6.3: Modelling forecasting, p. 236 and Box 8.2: Retail Geovisualization:
remoteness in mountain areas, p. 180.) 3D and animation for retail decision support, p. 262.)

xxxii About the authors

Jim Hogg graduated from the Dr Isaac Bonsu Karikari is the
University of Glasgow in 1968 with a Lands Commission’s Team Leader of
BSc Honours degree in Geography. the World Bank’s supported Land
He became a demonstrator and studied Administration Project in Ghana.
for an MSc degree in Aerial Photo- He is also the Head of Geographic
Interpretation in Geomorphology at Information Systems (GIS), Training
ITC Delft, Holland for a year before accepting an and Manpower Development Unit of the
appointment as research assistant in Remote Sensing Commission. He is an Associate Member of the
in the School of Forestry, University of Melbourne, Ghana Institution of Surveyors where he is the Chair
Victoria, Australia where he completed a PhD on of the Research and Development (R&D) Sub-
multispectral remote sensing in forestry. He went Committee and a Member of the Board of Examiners
on to become a lecturer in Geography at the (BOE) of the General Practice Division of the
University of Leeds in 1972 where he lectures in Institution. He holds a PhD from the University of
remote sensing, digital image processing, geographi- Leeds, UK, having obtained his MSc from the
cal information system and their potential for International Institute for Geo-information Science
research in physical geography and environmental and Earth Observation – ITC, Netherlands where
research. He has recently completed work on the EU he studied Geoinformation Systems for Urban
FIREGUARD project, which involved the use of Applications. (See Box 4.4: Encroachments on public
QuickBird and Ikonos stereo very high resolution land in Accra, Ghana, p. 122.)
images for measuring and assessing forest fuel loads
in Mediterranean countries of Europe. (See Box 5.4: Richard Kingston is a lecturer in
Fireguard using GIS and remote sensing for fire Urban Planning and GIS in the School of
management, p. 146.) Environment and Development at the University of
Manchester. He previously worked for Steve Carver
Born and raised in the north-east of as a research associate in the Centre for
England, Joseph Holden completed Computational Geography at the University of Leeds
an undergraduate degree in developing innovative PPGIS for local, regional and
Geography at the University of national decision-making problems. His research
Cambridge, followed by a PhD in focus is on developing, testing and implementing
peatland hydrology at the University Planning Support Systems for public participation in
of Durham. Joseph was recently a NERC Research spatial planning. He has published widely on PPGIS
Fellow at the University of Leeds and is now a lec- in academics journals and books. (See Box 8.7: Public
turer there where he works primarily in wetland participation in GIS, p. 274.)
environments but with more general interests in
hillslope hydrology, geomorphology and biogeo- Stuart Lane is a Professor in the
chemical cycling. He has over 30 journal Department of Geography at the University of
publications and another 30 published books, book Durham where he specialises in quantitative
chapters and reports. He is editor of an important approaches to the understanding of hydraulic and
new textbook titled ‘An introduction to physical hydrological processes. In particular, he specialises in
geography and the environment’ published by quantification of complex geomorphological sur-
Pearson Education in 2005. He sits on a number of faces, notably river systems, using remote sensing
international grant and journal review panels, is methods. He has worked in New Zealand, the Alps
leader of the River Basin Processes and Management and the UK both on research
research cluster and is Director of both the MSc in and consultancy contracts. (See Box 2.8: High resolu-
Catchment Dynamics and Management and the tion data solutions to complex modelling problems,
MRes in Hydrology at the University of Leeds. (See p. 60.)
Box 2.8: High resolution data solutions to complex
modelling problems, p. 60.)

About the authors xxxiii

Sarah Lindley is Lecturer in Prof. Jonathan Raper is Head of
Geographical Information Systems the Department of Information
in the School of Environment Science and founder of the giCentre
and Development, University of at City University, where he leads a
Manchester. She is a multi-discipli- group of 19 academics, researchers
nary researcher with particular and postgrads working on, among
interests in Geographical Information Systems and other things, location-based services, augmented
Science; air quality management; and sustainable reality, geographic data mining, geovisualisation
development. She has over ten years’ experience in and terrain and cityscape modelling. Jonathan is a
the development of spatially resolved emissions member of the Steering Committee of the Location
inventories and the wider use of Geographical and Timing Knowledge Transfer Network and
Information Systems and Science for air quality Editor in Chief of the Journal of LBS, newly launched
management applications. Sarah has been a member by Taylor and Francis. (See Box 3.8: Monitoring
of the UK Department of the Environment, Food coastal change with GIS, p. 102.)
and Rural Affairs (DEFRA) Air Quality Expert
Group since 2002. She also coordinates the data John Stillwell is Professor of
management activities for the EPSRC/UKCIP Migration and Regional Development in the School of
Building Knowledge for a Changing Climate Geography at the University of Leeds. Currently he is
research programme. (See Box 7.3: Using GIS to help also Director of the Census Interaction Data Service
manage air quality through estimating emissions of (CIDS) which is funded under the ESRC Census
atmospheric pollutants, p. 227.) Programme, and Coordinator of the ESRC initiative
on ‘Understanding Population Trends and Processes’.
Ross Purves is a lecturer in the GIS His research is primarily in the fields of migration stud-
Division of the Department of Geography at the ies and applied spatial analysis. (See Box 4.6: Delivering
University of Zurich. Prior to coming to Zurich, he census interaction data: WICID, p. 125.)
worked in the Geography Deaprtment of the
University of Edinburgh. His research in GIScience Jo Wood is a Senior Lecturer in
covers a range of themes, from Geographic Geographic Information at the giCentre, City
Information Retrieval to the environmental model- University, London. He has been involved in teach-
ling of processes related to snow and ice and, in ing and research in GI Science since 1990. His
particular, issues relating to the uncertainty in mod- research interests are in terrain modelling, visualiza-
elling of such processes. (See Box 10.7: Uncertainty tion of surfaces, object-oriented modelling of GI and
and terrain data in ice-sheet modelling, p. 321.) collaborative networks in GI. He is author of the
widely used GIS LandSerf, the text Java Programming for
Spatial Sciences and is course director of the Masters in
Geographic Information (MGI). When not analyz-
ing terrain with a GIS, he can usually be found
walking or cycling across it. (See Box 6.9: Identifying
mountains with GIS, p. 207.)



A head in running heading xxxv

xxxvi Chapter 5 Data input and editing

PART1Fundamentals
of GIS

CHAPTER1

What is GIS?

Learning outcomes

By the end of this chapter you should be able to:
I Explain what GIS is
I Give examples of the applications of GIS
I Outline the characteristics of GIS
I Describe how the real world is represented in GIS
I Provide examples of the analysis GIS can perform
I Know where to find more information about GIS

Introduction 3

I INTRODUCTION I Implications. If I move to a new home in this
location, how far will I be from the office, gym or
Every day you ask questions with a spatial compo- coffee shop? If we build a new theme park here,
nent. Whether you are at work, studying or at what will be the effect on traffic flows? What would
leisure you probably ask spatial questions. Many of be the time saving if we delivered our parcels using
these questions you answer for yourself without ref- this route, rather than an alternative?
erence to a map or a GIS, but both of these tools
could help. GIS has particular value when you need This chapter provides an overview of GIS. It exam-
to answer questions about location, patterns, trends ines what GIS is, what it can do and, in brief, how it
and conditions such as those below: works. We begin with a look at the types of generic
questions GIS can answer and expand on these with
I Location. Where is the nearest bookshop? Where reference to case studies. GIS is then defined, and a
are stone age settlements located in Europe? range of issues and ideas associated with its use iden-
Where are areas of forestry in which Norwegian tified. Much of the material introduced in this
Spruce trees can be found? chapter will be covered in more detail later in the
book, and pointers to the appropriate sections are
I Patterns. Where do high concentrations of students provided. If we return to the car analogy introduced
live in this city? What is the flow of traffic along in the Preface, this is where you find out what
this motorway? What is the distribution of crime exactly a car is, why it is useful and what you need to
incidents in London? know to make it work.

I Trends. How are patterns of retailing changing in The generic questions that a GIS can help to
response to the development of out-of-town answer can be summarized as:
superstores? Where have glaciers retreated in the
European Alps? Where have changes to the I Where are particular features found?
population of polar bears occurred?
I What geographical patterns exist?
I Conditions. Where can I find holiday
accommodation that is within 1 km of a I Where have changes occurred over a given period?
wind surfing beach and accessible by public
transport? Where is there flat land within 500 m I Where do certain conditions apply?
of a major highway? Where are there over
100,000 potential customers within a 5 mile I What will the spatial implications be if an
radius of a railway station? organization takes certain action?

Box 1.1 provides examples of problems that can be
addressed by asking these types of questions in an
imaginary ski resort called Happy Valley.

BOX 1.1 Questions GIS CASE
could answer in Happy STUDY
Valley

The Happy Valley GIS has been established to help GIS is used to produce maps of the ski area. Visitors
the ski resort’s managers improve the quality of the can also ask direct questions about the location of
ski experience for visitors. The examples below are facilities using ‘touch screen’ computerized informa-
only a few of the situations in which asking a spatial tion points located in shops and cafes throughout the
question can help with the management of Happy ski resort. These information points provide skiers
Valley. You will find many other examples throughout with a customized map showing them how to find the
the rest of this book. facilities they require.

1 Where are particular features found? When skiers 2 What geographical patterns exist? Over the last
visit Happy Valley they need to know where all the vis- two ski seasons there have been a number of
itor facilities are located. To help, the Happy Valley
accidents involving skiers. All these incidents  

4 Chapter 1 What is GIS?

BOX 1.1 ment team provides information on which ski pistes
Figure 1.1 Facilities in Happy Valley ski resort are open. Since this depends on the snow cover, ava-
lanche danger and wind strength, data on these
factors are regularly added to the GIS from reports
provided by the ski patrols and local weather service.
The warden can use this information and the GIS to
help identify which runs should be opened or closed.

5 What will the spatial implications be if an organi-
zation takes certain action? The access road to Happy
Valley is now too narrow for the number of skiers
visiting the area. A plan is being prepared for
widening the road. However, any road-widening
scheme will have impacts on a local nature reserve
as well as surrounding farm land. The Happy Valley
GIS is being used to establish the amount of land
that is likely to be affected under different road-
widening schemes.

Dataset and activities relating to the Happy Valley
Case Study can be found online at www.pearsoned.
co.uk/heywood

have been located and entered into the GIS. The Happy Figure 1.2 Avalanche! (Source: Richard Armstrong)
Valley management team is trying to establish
whether there is any common theme or spatial pattern
to the accidents. Do accidents of a certain type occur
only on specific ski pistes, at certain points on a ski
piste such as the lift stations, or at particular times of
day? So far one accident black spot has been identified
where an advanced ski run cuts across a slope used by
beginners, just below a mountain restaurant.

3 Where have changes occurred over a given time
period? In Happy Valley avalanches present a danger
to skiers who wish to venture off the groomed ski
pistes. The management team and the ski patrol use
the GIS to build up a picture of snow cover throughout
the area. This is done by regularly recording snow
depth, surface temperature, snow water content and
snow strength at a number of locations. A study of the
geographical changes in these parameters helps the
management team prepare avalanche forecasts for
different locations in Happy Valley.

4 Where do certain conditions apply? Every day,
during the winter season, the Happy Valley manage-

If you have a geographical background you may and others using spatial data have been unable to
be asking what is new about these generic questions. find answers to their questions because of the
Are these not the questions that geographers have volume of data required and a lack of time and tech-
been contemplating and answering for centuries? In niques available to process these data. The following
part they are, though in many cases geographers examples of GIS applications are used to illustrate

Introduction 5

the capabilities of GIS as a tool for geographical or a new airport. Sometimes the task is more
analysis. All involve the manipulation of data in demanding than others, involving searches through
ways that would be difficult or impossible by hand, large numbers of maps and related documents.
and each illustrates different issues associated with
the application of GIS. Radioactive waste is waste material that contains
levels of ionizing radiation considered harmful to
Searching for sites health. It is generated by the nuclear industry,
nuclear power generation, weapons manufacture,
Searching for the optimum location to put some- medical and research establishments (Figure 1.3).
thing is a task performed by individuals and Radioactive waste is categorized according to the
organizations on a regular basis. The task may be to level of radiation emitted and the length of time for
find a site for a new retail outlet, a new oil terminal which it will be radioactive. Careful management is
required and ultimately governments and the

(a)

(b) (c) (d)

Figure 1.3 A nuclear power station (a) and nuclear waste containers (b, c and d)
(Sources: (a) courtesy of author; (b) Getty Images; (c,d) British Nuclear Fuels Ltd)

6 Chapter 1 What is GIS?

Figure 1.4 Anti-nuclear protestors

nuclear industry have to find appropriate disposal NIREX. Where, therefore, should NIREX site a
options. Over the last 20 years finding a suitable site nuclear waste repository?
for the disposal of radioactive waste in the UK has
become a sensitive and important issue. NIREX In the past, NIREX used a pen and paper
(Nuclear Industry Radioactive Waste Executive) is approach to sieve through large numbers of paper
the UK company with responsibility for the identifi- maps containing data about geology, land use,
cation of suitable radioactive waste disposal sites. land ownership, protected areas, population and
What to do with the UK’s radioactive waste is cur- other relevant factors. Areas of interest were
rently being reviewed by the Committee on traced from these maps by hand, then the trac-
Radioactive Waste Management (CoRWM). In the ings were overlaid to identify areas where
meantime NIREX has the task of interpreting cur- conditions overlapped. NIREX is not the only
rent government radioactive waste policy and siting organization that has searched for sites using
guidelines and presenting possible sites at public these techniques. This was a standard approach
inquiries. One of the problems for NIREX is the lack employed in the siting of a wide range of activities
of comprehensive and coherent guidelines for the including shopping centres, roads and offices
identification of suitable sites. Another is that (Figure 1.5). The method is time-consuming and
radioactive waste is a strong political issue because means that it is impossible to perform the analysis
nobody wants a disposal facility in their neighbour- for more than a few different siting criteria. The
hood and protests against potential sites are best sites are often missed. GIS techniques offer an
common (Figure 1.4). However, NIREX is expected alternative approach, allowing quick remodelling
to show that it has followed a rational procedure for for slight changes in siting criteria, and produce
site identification (Department of the Environment, results as maps eminently suitable for presenta-
1985). Hydrology, population distribution and acces- tion at public inquiries.
sibility are examples of important siting factors, but
how such factors should be interpreted is left up to Openshaw et al. (1989) first demonstrated the use
of GIS for this application, and their method is sum-
marized in Figure 1.6. They established a number of

Introduction 7

(a) Step one nature conservation areas and population statistics).
These were converted from paper to digital format
by digitizing (this technique is explained in Chapter
5), or acquired from existing digital sources such as
the UK Census of Population. These data layers were
then processed so that they represented specific
siting criteria. The geology layer was refined so that
only those areas with suitable geology remained; the
transport layer altered so that only those areas close
to major routes were identified; and the nature con-
servation layer processed to show protected areas
where no development is permitted. The population
layer was analyzed so that areas with high popula-
tion densities were removed. The four data layers are
shown in Figure 1.7.

GIS software was then used to combine these
new data layers with additional layers of informa-
tion representing other siting criteria. The final

Identify relevant
siting factors

Collect appropriate
data and digitize

(b) Step two Yes

Identify siting No New
criteria and map factors required?

Overlay

Examine No
output
(c) Step three
Figure 1.5 Tracing paper overlay OK?

Yes

Final potential
areas

data layers, each containing data for a separate siting Figure 1.6 Using GIS for siting a NIREX waste site
criterion (for example, geology, transport networks, (Source: Adapted from Openshaw et al., 1989)

8 Chapter 1 What is GIS?

(a) Geology (b) Population

(c) Transport (d) Conservation criterion

Figure 1.7 Radioactive waste case study: geology, population, transport and conservation criteria maps.
(Sources: (a) British Geological Survey. © NERC, IPR/71-39C reproduced by permission; (b) Office for National
Statistics (www.statistics.gov.uk). Crown copyright material is reproduced with permission of the controller of
HMSO; (c) Ordnance Survey; (d) Joint Nature Conservation Committee (www.jncc.gov.uk))

Introduction 9

(a) (b)

Figure 1.8 Radioactive waste case study (a and b) results from different siting scenarios

result was a map showing the locations where all Evaluating land use planning
the specified siting criteria were satisfied and, thus,
a number of locations suitable for the siting of a Virtually every country in the world has areas of
nuclear waste repository. The advantage of using natural beauty and conservation value that are
the GIS to perform this task was that the siting cri- managed and protected in the public interest (Figure
teria could be altered and the procedure repeated 1.9). Those managing these areas face the problem of
with relative ease. Examples for several siting sce- balancing human activities (such as farming, indus-
narios are shown in Figure 1.8. These illustrate how try and tourism) with the natural elements of the
changes in siting criteria influence the geographical landscape (such as climate, flora and fauna) in order
distribution of potential sites. to maintain the special landscape character without
exploitation or stagnation.
This example shows how a GIS approach allows
comparative re-evaluation and testing of data and The protected area of Zdarske Vrchy, in the
conditions. In this way the decision maker can eval- Bohemian–Moravian highlands of the Czech
uate options in a detailed and scientific manner. The Republic, is an example of an area that has suf-
work of Openshaw et al. (1989) also illustrates three fered as a result of ill-considered state control.
other important issues associated with the use of Unregulated farming, tourism and industrial activi-
GIS: the problem of errors in spatial data sets; the ties have placed the landscape under severe
difficulty in establishing criteria for abstract spatial pressure. Czech scientists and environmental man-
concepts; and the potential value of using GIS to agers have relied on traditional mapping and
communicate ideas (Box 1.2). statistical techniques to monitor, evaluate and pre-

10 Chapter 1 What is GIS?

BOX 1.2 Issues raised by CASE
the NIREX case study STUDY

I Errors in source data, such as those introduced tance (for example, 10, 20 or 30 km) ‘far from’ rep-
during the conversion of data to digital form, may resents. In some cases rules may be applied to
have a significant effect on the GIS site-searching guide this process, but in others the numerical rep-
process. Mistakes in capturing areas of appropriate resentation of a criterion may depend upon the
geology from paper maps may lead to inappropriate preferences of the person responsible for choosing
waste repository sites being identified, because and implementing the criterion.
areas on the ground will have different geological I GIS output can be used to inform public participation
properties from those recorded in the GIS. Errors in in the decision-making process. A series of maps
spatial data sets and the associated issues of data could be used to illustrate why a particular geo-
quality are discussed in detail in Chapter 10. graphical location has been identified as a suitable
site for the disposal of radioactive waste. However,
I The GIS site-searching process relies on the trans- the issues raised above – data quality and the prob-
lation of abstract (or ‘fuzzy’) concepts such as ‘near lems of creating spatial criteria for abstract concepts
to’ and ‘far from’ into precise conditions that can be – suggest that output from GIS should be viewed with
mapped. This can be a problem. How do you create caution. Just because a map is computer-generated
a map that shows all the geographical zones ‘far does not mean that the picture it presents is correct.
from’ a centre of population? The only method is to More on this issue can be found in Chapter 8.
make an arbitrary decision about what sort of dis-

dict the consequences of this exploitation. ing programmes, socio-economic surveys and
However, with the change in the political adminis- tourism studies have been mapped and overlaid to
tration of the country there are scientists and identify areas of compatibility and conflict.
managers who are looking not only into a policy of
sustainable development for Zdarske Vrchy, but In this context GIS permits scientists and man-
also at GIS as a tool to help with policy formulation agers in Zdarske Vrchy to interact with their data
(Petch et al., 1995). and ask questions such as:

GIS is seen as a tool to bring together disparate I What will be the long-term consequences of
data and information about the character and activi- continuing recreational activity for the
ties that take place in the Zdarske Vrchy region. landscape? (Figure 1.10)
Data from maps, aerial photographs, satellite
images, ecological field projects, pollution monitor- I Where will damage from acid rain occur if a
particular industrial plant continues to operate?

Figure 1.9 Zdarske Vrchy

Introduction 11

(a) Ecological survey (d) Village population

(b) Relief (20 m contours) (e) Recreational load: this map has been produced by
combining the maps shown in panels a–d with several
other data layers for the region

(c) Geology and hydrology I Where should landscape conservation zones be
established?
Figure 1.10 GIS landscape assessment data layers:
Zdarske Vrchy (after Downey et al., 1991, Downey et al., 1992 and
Petch et al., 1995).
I Where is the best location for the re-introduction
of certain bird species? One application of GIS in Zdarske Vrchy has been to
identify areas of the landscape for conservation.
Traditionally, water storage in the region has relied on
the use of natural water reservoirs such as peat wet-
lands and old river meanders (Petch et al., 1995).
Current land use practices, in particular forestry and
farming, are resulting in the gradual disappearance of
these features. The consequences of these changes have
been localized droughts and floods in areas further
downstream. In turn, these changes have brought
about a reduction in plant species and wildlife habitats.
Managers in the Zdarske Vrchy region wanted to

12 Chapter 1 What is GIS?

Remote Identify Paper
sensing relevant ecological maps
from field work
data data

Topographic
maps

Digitize

Extract Digitize
land cover ecological

maps zones

Extracted from Overlay to Extracted from
water management identify sensitive nature conservation

database zones database

Rivers Ecologically Nature
and sensitive conservation
zones
ponds sites

Overlay to
identify sensitive zones

for consideration as
conservation zones

Decide No
which zones to

implement

Yes

New
conservation
zone established

Figure 1.11 Using GIS for identifying conservation zones in Zdarske Vrchy (Source: Adapted from Petch et al., 1995)

identify conservation zones to protect the remaining I the type of land use, as certain agricultural
natural water reservoirs as well as to identify those practices exploit the water retention capacity of
areas where it may be possible to restore the water the landscape; and
retention character of the landscape. To do this they
needed to establish the characteristics of the landscape I the presence or absence of human-enhanced
that determine whether or not a particular location is water drainage channels.
likely to retain water. Specialists in hydrology, geology
and ecology were consulted to identify a range of GIS professionals were then asked to find appropri-
important criteria describing: ate sources of spatial data that could be used to
represent these criteria. A range of sources was iden-
I the type of soil and its water retention ability; tified including:

I the character of the topography (for example, I paper maps (for soil type and geology);
presence or absence of hollows or hills);
I contour maps (for topography);