Dynamic land use/cover change modelling: Geosimulation and multiagent-based modelling

152 Pages·2012·6.015 MB·English

by  Jamal Jokar Arsanjani (auth.)

Springer Theses
Recognizing Outstanding Ph.D. Research
For furthervolumes:
http://www.springer.com/series/8790
Aims and Scope
The series ‘‘Springer Theses’’ brings together a selection of the very best Ph.D.
theses from around the world and across the physical sciences. Nominated and
endorsed by two recognized specialists, each published volume has been selected
for its scientific excellence and the high impact of its contents for the pertinent
fieldofresearch.Forgreateraccessibilitytonon-specialists,thepublishedversions
includeanextendedintroduction,aswellasaforewordbythestudent’ssupervisor
explaining the special relevance of the work for the field. As a whole, the series
will provide a valuable resource both for newcomers to the research fields
described, and for other scientists seeking detailed background information on
specialquestions.Finally,itprovidesanaccrediteddocumentationofthevaluable
contributions made by today’s younger generation of scientists.
Theses are accepted into the series by invited nomination only
and must fulfill all of the following criteria
• They must be written in good English.
• The topic should fall within the confines of Chemistry, Physics and related
interdisciplinary fields such as Materials, Nanoscience, Chemical Engineering,
Complex Systems and Biophysics.
• The work reported in the thesis must represent a significant scientific advance.
• Ifthethesisincludespreviouslypublishedmaterial,permissiontoreproducethis
must be gained from the respective copyright holder.
• They must have been examined and passed during the 12 months prior to
nomination.
• Each thesis should include a foreword by the supervisor outlining the signifi-
cance of its content.
• The theses should have a clearly defined structure including an introduction
accessible to scientists not expert in that particular field.
Jamal Jokar Arsanjani
Dynamic Land-Use/Cover
Change Simulation:
Geosimulation and Multi
Agent-Based Modelling
Doctoral Thesis accepted by
University of Vienna, Austria
123
Author Supervisor
Dr. JamalJokarArsanjani Prof.Dr. WolfgangKainz
Department of Geography Department of Geography
and Regional Research and Regional Research
Universityof Vienna Universityof Vienna
Universitätsstraße 7 Universitätsstraße 7
A-1010Vienna A-1010Vienna
Austria Austria
e-mail: [email protected]
ISSN 2190-5053 e-ISSN 2190-5061
ISBN 978-3-642-23704-1 e-ISBN978-3-642-23705-8
DOI 10.1007/978-3-642-23705-8
SpringerHeidelbergDordrechtLondonNewYork
LibraryofCongressControlNumber:2011937768
(cid:2)Springer-VerlagBerlinHeidelberg2012
Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis
concerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation,broadcasting,
reproductiononmicrofilmorinanyotherway,andstorageindatabanks.Duplicationofthispublication
orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,
1965,initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violations
areliabletoprosecutionundertheGermanCopyrightLaw.
Theuseofgeneraldescriptivenames,registerednames,trademarks,etc.inthispublicationdoesnot
imply, even in the absence of a specific statement, that such names are exempt from the relevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
Coverdesign:eStudioCalamar,Berlin/Figueres
Printedonacid-freepaper
SpringerispartofSpringerScience+BusinessMedia(www.springer.com)
Parts of this thesis have been published in the following journal articles:
J. Jokar Arsanjani, W. Kainz, Integration Of Spatial Agents And Markov
Chain Model in Simulation of Urban Sprawl, In Proceeding of AGILE
conference 2011, Utrecht, the Netherlands (peer reviewed)
J. Jokar Arsanjani, M. Helbich, W. Kainz, A. Darvishi B., Integration of
LogisticRegressionandMarkovChainModelstoSimulateUrbanExpansion,
Submitted to the International Journal of Applied Earth Observation and
Geoinformation, 2011 (Accepted for publication)
J. Jokar Arsanjani, W. Kainz, A. Mousivand, Tracking Dynamic Land Use
ChangeUsingSpatiallyExplicitMarkovChainBasedonCellularAutomata-
the Case of Tehran, International Journal of Image and Data Fusion, 2011
(In press)
J. Jokar Arsanjani, W. Kainz, M. Azadbakht, Monitoring and Geospatially
ExplicitSimulation ofLand Use Dynamics: fromCellular Automata towards
Geosimulation—CaseStudyTehran,Iran,InProceedingofISDIF2011,China
J. Jokar Arsanjani, M. Helbich, W. Kainz, The Emergence of Urban Sprawl
PatternsinTehranMetropolisthroughAgentBasedModelling,inpreparation
Supervisor’s Foreword
Land use and land cover change are two subjects that have triggered a large
number of research activities and resulted in a wealth of different approaches to
detect past change and also to predict future development. Among the most
prominentmethodsarethosethatuseremotesensingandimageanalysiscombined
withvariousstatisticalandanalyticalprocedures.Theyallrequireaseriesofdata
over longer periods, appropriate land use maps, and related information. It is not
always easy to acquire or access these data due to a simple lack of data or
administrativeaccessrestrictions.Itisthereforeimperativetomakeuseofsatellite
data and other easier accessible data of reasonable resolution.
Many large cities face pressing problems with—sometimes uncontrolled—
growth and sprawl, in particular when their expansion is limited by natural and
otherconditions.Tehranisoneofthesecitieswhoseexpansionisafact,butwhich
also experiences severe topographic constraints by its location at the foothills of
theAlborzMountains.Tehranisaverydynamiccitywhichgrewrapidlyoverthe
last decades. Being an Iranian it was therefore very logical for Dr. Jamal Jokar
Arsanjanitochoosethecapitalofhishomecountryasastudyareaandatthesame
time a city that has to cope with all the problems of urban sprawl.
TheoriginalfocusofDr.JokarArsanjani’sworkisonagent-basedmodelingto
predictlandcoverchangefortheTehranarea.Thisalonewouldalreadyhavebeen
aninterestingendeavorworthinvestigating.However,arealvalueoftheworklies
also inthe extensive application andcomparison of traditional methods topredict
land cover change. These methods are cellular automata, Markov chain model,
cellular automata Markov model, and the hybrid logistic regression model. In his
thesis all these methods have been applied to the Tehran area to analyze and
predict land cover change. In this respect the work can also serve as a text
explainingthedifferentapproachesintheirtheoreticalcharacteristicsandpractical
applications.Itisaparticularvaluethattheadvantagesanddisadvantagesofthese
methods are clearly exposed and explained.
Based on the preliminary findings of the different methods, finally, an agent-
based model was developed that consist of government agents, developer agents,
and resident agents, in order to simulate land cover change. Various parameters
vii
viii Supervisor’sForeword
and behaviors were modeled and programmed in the ArcGIS environment.
Since almost nothing in the real world follows a crisp classification, many tradi-
tional approaches suffer from a lack of adequately representing the real world
situation.Fuzzylogicisonewaytointroduceuncertaintyandvaguenesstospatial
analysis. Dr. Jamal Jokar Arsanjani uses fuzzy membership functions for the
relevantfactorsinhisgeo-simulationresearchtorepresentamorenaturalbehavior
of the agents. This offers a more realistic analysis and provides results that better
suit a real world situation.
The major value of this work is twofold: it shows a detailed comparison of
existingmethodsforlandcoverchangemodeling,anditpresentsanovelapproach
ingeo-simulationbyapplyingagent-basedmodelinginafuzzysetting.Thethesis
has already spawned several journal papers and Dr. Jokar Arsanjani’s approach
opens new perspectives for scientific problems in environmental monitoring,
modeling and change detection.
Vienna, June 2011 Prof. Dr. Wolfgang Kainz
Contents
1 General Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Problem Statement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.1 Rapid Urban Expansion of Tehran. . . . . . . . . . . . . . 3
1.2.2 Limitations of Previous Approaches. . . . . . . . . . . . . 4
1.3 Research Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.5 Research Objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.6 Research Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.7 Organisation of the Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . 7
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Land Use/Cover Change. . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Land Use/Cover Change Causes and Consequences. . . . . . . . . 10
2.3.1 Loss of Biodiversity. . . . . . . . . . . . . . . . . . . . . . . . 10
2.3.2 Climate Change. . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.3 Pollution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.4 Other Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Driving Forces of the Land Use/Cover Changes . . . . . . . . . . . 11
2.5 Land Use/Cover Change Simulation. . . . . . . . . . . . . . . . . . . . 12
2.6 Land Use Change Trend. . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.7 Predicting Future Land Use Patterns. . . . . . . . . . . . . . . . . . . . 14
2.8 Simulating Sprawl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.9 Approaches to the LUCC Modelling . . . . . . . . . . . . . . . . . . . 15
2.10 Agent-Based Modelling and Geosimulation Terminology . . . . . 15
2.10.1 Agents and Agent-Based Models . . . . . . . . . . . . . . . 16
ix
x Contents
2.11 Characteristics of the Geosimulation Model . . . . . . . . . . . . . . 18
2.11.1 Management of Spatial Entities . . . . . . . . . . . . . . . . 18
2.11.2 Management of Spatial Relationships. . . . . . . . . . . . 19
2.11.3 Management of Time . . . . . . . . . . . . . . . . . . . . . . . 19
2.11.4 Direct Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.12 The Basic of Geosimulation Framework: Automata. . . . . . . . . 20
2.13 Cellular Automata versus Multi-Agent Systems. . . . . . . . . . . . 20
2.14 Geographic Automata System . . . . . . . . . . . . . . . . . . . . . . . . 21
2.14.1 Definitions of Geographic Automata Systems . . . . . . 21
2.14.2 Geographic Automata Types . . . . . . . . . . . . . . . . . . 22
2.14.3 Geographic Automata States and State
Transition Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.14.4 Geographic Automata Spatial Migration Rules. . . . . . 23
2.14.5 Geographic Automata Neighbours
and Neighbourhood Rules. . . . . . . . . . . . . . . . . . . . 23
2.14.6 Types of Simulation Systems for
Agent-Based Modelling. . . . . . . . . . . . . . . . . . . . . . 24
2.15 Current Simulation Systems . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.15.1 ASCAPE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.15.2 StarLogo. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.15.3 NetLogo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.15.4 OBEUS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.15.5 AgentSheets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.15.6 AnyLogic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.15.7 SWARM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.15.8 MASON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.15.9 NetLogo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.15.10 Repast. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.15.11 Agent Analyst Extension. . . . . . . . . . . . . . . . . . . . . 28
2.16 Selection of ABM Implementation Toolkit . . . . . . . . . . . . . . . 29
2.17 Designing a Multi Agent System. . . . . . . . . . . . . . . . . . . . . . 29
2.18 Fuzzy Decision Theory in Geographical Entities . . . . . . . . . . . 31
2.18.1 Fuzzy Geographical Entities . . . . . . . . . . . . . . . . . . 33
2.18.2 Processing Fuzzy Geographical Entities . . . . . . . . . . 34
2.19 The Analytical Hierarchy Process Weighting. . . . . . . . . . . . . . 35
2.20 Moran’s Autocorrelation Coefficient Analysis. . . . . . . . . . . . . 36
2.21 Accuracy Assessment and Uncertainty in Maps Comparison. . . 37
2.21.1 Calibration and Validation. . . . . . . . . . . . . . . . . . . . 37
2.21.2 Techniques of Validation for Land
Change Models . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.22 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40