Living Rev Landscape Res 3 (2009) 2httpwwwlivingreviewsorglrlr-2009-2 in landscape research

RRRRLLLLL I V I N G REVIEWS

Simulation Models on HumanndashNature Interactions in Urban

Landscapes A Review Including Spatial Economics System

Dynamics Cellular Automata and Agent-based Approaches

Dagmar HaaseHelmholtz Centre for Environmental Research - UFZDepartment of Computational Landscape Ecology

Permoserstr 15 D-04318 Leipzig Germanyemail dagmarhaaseufzde

httpwwwufzdeindexphpen=4576

Nina SchwarzHelmholtz Centre for Environmental Research - UFZDepartment of Computational Landscape Ecology

Permoserstr 15 D-04318 Leipzig Germanyemail ninaschwarzufzde

httpwwwufzdeindexphpen=13323

Living Reviews in Landscape ResearchISSN 1863-7329

Accepted on 2 April 2009Published on 9 April 2009

Abstract

Urbanisation belongs to the most complex and dynamic processes of land use and land-scape change At present we claim ldquothe millennium of the citiesrdquo since more than half ofthe currently 66 billion world population is living in urban areas Due to the huge impact ofurban land consumption on environment and landscape this paper provides a review of exist-ing urban land use models The review analyses non-spatially explicit economic and systemdynamics models spatially explicit cellular automata and agent-based model approaches byaddressing the respective conceptual approach model components and causal relationshipsincluding feedbacks Based upon the review conclusions are drawn regarding the future devel-opment of urban landscape models as well as on indispensable causal relationships and theirrepresentation when modelling urban systems

Keywords urban landscape simulation models land use change review system dynamicscellular automata agent-based model feedback causalities

This review is licensed under a Creative CommonsAttribution-Non-Commercial-NoDerivs 30 Germany Licensehttpcreativecommonsorglicensesby-nc-nd30de

Imprint Terms of Use

Living Reviews in Landscape Research is a peer reviewed open access journal published by theLeibniz Centre for Agricultural Landscape Research (ZALF) Eberswalder Straszlige 84 15374Muncheberg Germany ISSN 1863-7329

This review is licensed under a Creative Commons Attribution-Non-Commercial-NoDerivs 30Germany License httpcreativecommonsorglicensesby-nc-nd30de

Because a Living Reviews article can evolve over time we recommend to cite the article as follows

Dagmar Haase and Nina SchwarzldquoSimulation Models on HumanndashNature Interactions in Urban Landscapes A ReviewIncluding Spatial Economics System Dynamics Cellular Automata and Agent-based

ApproachesrdquoLiving Rev Landscape Res 3 (2009) 2 [Online Article] cited [ltdategt]

httpwwwlivingreviewsorglrlr-2009-2

The date given as ltdategt then uniquely identifies the version of the article you are referring to

Article Revisions

Living Reviews supports two different ways to keep its articles up-to-date

Fast-track revision A fast-track revision provides the author with the opportunity to add shortnotices of current research results trends and developments or important publications tothe article A fast-track revision is refereed by the responsible subject editor If an articlehas undergone a fast-track revision a summary of changes will be listed here

Major update A major update will include substantial changes and additions and is subject tofull external refereeing It is published with a new publication number

For detailed documentation of an articlersquos evolution please refer always to the history documentof the articlersquos online version at httpwwwlivingreviewsorglrlr-2009-2

Contents

1 Introduction 511 Urbanisation of landscapes 512 The ldquoidealrdquo urban land use model 513 Existing reviews on urban land use models 614 The purpose of this review 7

2 Evaluating urban land use simulation models 8

3 Models under review 10

4 Representation of urban landscapes 1441 Spidergrams 1542 Relationships between human sphere and land use 1543 Impact of land use on environment 1544 Feedback from environment to human sphere 1845 Feedbacks between local and regional scale 18

5 Conclusions 19

6 Acknowledgements 20

7 Annex 20

References 40

List of Tables

1 Overview of main purposes and components investigated in reviewed models 112 Main components of urban systems ndash do the models under review include them 143 Household life cycle model for residential relocation behaviour [SE 1] 214 Simulation of demographic changes and the housing market [SE 2] 225 A System Dynamics Approach to Land Use Transportation System Performance

Modeling [SD 2] 236 CLUE-s (Conversion of Land Use and its Effects) [CA 1] 247 CUF-2 (California Urban Futures) [CA 2] 258 CURBA (California Urban and Biodiversity Analysis)(CA 3] 269 ILUMASS (Integrated Land-Use Modelling and Transportation System Simulation)

[ABM 1] 2710 ILUTE (Integrated Land Use Transportation Environment model) [ABM 2] 2811 Modelling biodiversity and land use [SD 3] 2912 MOLAND [CA 4] 3013 PUMA (Predicting Urbanisation with Multi-Agents) [ABM 3] 3114 Rotterdam urban dynamics [SD 4] 3215 SCOPE (South Coast Outlook and Participation Experience) [SD 5] 3316 Simulation of polycentric urban growth dynamics through agents [ABM 4] 3417 SLEUTH (Slope Landuse Exclusion Urban Extend Transportation and Hillshade)

[CA 5] 3518 Urban dynamics [SD 1] 3619 Urban transformation process in the Haaglanden region [SD 6] 37

20 Urban travel system [SD 7] 3821 UrbanSim [ABM 5] 39

Models on HumanndashNature Interactions 5

1 Introduction

11 Urbanisation of landscapes

Urbanisation is one of the most complex and dynamic processes of landscape change Althoughonly about 4 of the worldrsquos land area is urbanised and densly populated (Ramankutty et al2006) we claim ldquothe millennium of the citiesrdquo since more than half of the currently 66 billionworld population is living in urban areas (United Nations 2008 2009 PRB 2007 EEA 2006Kasanko et al 2006) Projections for the future show that urbanisation ndash in terms of an increasingshare of population living in urban areas ndash is very likely to continue (Batty et al 2003 EEA2006 Lutz et al 2001) Urbanisation is not only a societal problem but also an environmentalone because it contradicts a normative ideal of ldquoa natural or un-spoiled landscaperdquo in spatialplanning (Nuissl et al 2008) In a multitude of studies it has been shown that land consumption isusually detrimental to the environment in different regards (eg Johnson 2001 Antrop 2004) Itsimpact reduces the ability of landscapes to fulfil human requirements and thus impairs ecosystemservices and landscape functions in various ways (de Groot et al 2002 Millennium EcosystemAssessment 2005 Curran and de Sherbinin 2004) Individual ecosystem services and quality oflife aspects that are affected by urbanisation include the production of food the regulation ofenergy and matter flows water supply the provision of biodiversity and of health and recreationand the supply of green space and natural aesthetic values (Alberti 1999) Suburbanisation andurban sprawl were the dominating land consumption processes in North America and Europeafter WW II (Batty 2008) Recently high growth rates in developing countries have led toenormous environmental loads as discussed above (Heinrichs and Kabisch 2006) As urban systemsare very densely populated and their land use components highly interlinked (Liu et al 2007)developing views about their future is both a major concern in landscape research and a complextask Modelling land use relationships helps to understand underlying drivers of land use changeto create future land use scenarios and assess possible environmental impacts (Lambin and Geist2006 Ravetz 2000)

12 The ldquoidealrdquo urban land use model

A variety of land use change models particularly for urban landscapes already exist ranging fromspecific case studies to generic tools for a variety of urban regions These models differ largely interms of their structure their representation of both space and human decisions and their method-ological implementation Compared to land use change models in open landscapes urban areas areshaped particularly by human activities societal processes and humanndashnature interactions (Coucle-lis 1997) In addition to implemented simulation models a number of articles and book chapterselaborate on the ldquoidealrdquo integrated model theoretically necessary causal feedback loops etc Theseldquoidealrdquo models shall serve as analytical frameworks to better understand the systems under studyOften authors use frameworks like the DPSIR-framework (drivers pressures state impact re-sponses) of the European Environment Agency (EEA) to conceptualise these conceptual modelsAccording to Verburg ldquothe main drawback of using these analytical frameworks is the assumptionof one-directional processes between driving factors and impactsrdquo (Verburg 2006 p 1173) becausein reality it is difficult to differentiate between impacts and drivers in a system Burgi et al (2004)distinguish five major types of driving forces socioeconomic political technological natural andcultural Furthermore they differentiate between primary secondary and tertiary driving forcesas well as between intrinsic and extrinsic driving forces (Burgi et al 2004) In their introductionto urban simulation Waddell and Ulfarsson (2004) sketched urban markets and agents choicesand interactions in an ldquoidealrdquo urban land use model Timmermans (2006) criticizes that presenturban models focus on functional chains like the following demand causes allocation across spacewhich in turn causes traffic flows based upon which a transportation model calculates travel times

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

20 Urban travel system [SD 7] 3821 UrbanSim [ABM 5] 39

Models on HumanndashNature Interactions 5

1 Introduction

11 Urbanisation of landscapes

Urbanisation is one of the most complex and dynamic processes of landscape change Althoughonly about 4 of the worldrsquos land area is urbanised and densly populated (Ramankutty et al2006) we claim ldquothe millennium of the citiesrdquo since more than half of the currently 66 billionworld population is living in urban areas (United Nations 2008 2009 PRB 2007 EEA 2006Kasanko et al 2006) Projections for the future show that urbanisation ndash in terms of an increasingshare of population living in urban areas ndash is very likely to continue (Batty et al 2003 EEA2006 Lutz et al 2001) Urbanisation is not only a societal problem but also an environmentalone because it contradicts a normative ideal of ldquoa natural or un-spoiled landscaperdquo in spatialplanning (Nuissl et al 2008) In a multitude of studies it has been shown that land consumption isusually detrimental to the environment in different regards (eg Johnson 2001 Antrop 2004) Itsimpact reduces the ability of landscapes to fulfil human requirements and thus impairs ecosystemservices and landscape functions in various ways (de Groot et al 2002 Millennium EcosystemAssessment 2005 Curran and de Sherbinin 2004) Individual ecosystem services and quality oflife aspects that are affected by urbanisation include the production of food the regulation ofenergy and matter flows water supply the provision of biodiversity and of health and recreationand the supply of green space and natural aesthetic values (Alberti 1999) Suburbanisation andurban sprawl were the dominating land consumption processes in North America and Europeafter WW II (Batty 2008) Recently high growth rates in developing countries have led toenormous environmental loads as discussed above (Heinrichs and Kabisch 2006) As urban systemsare very densely populated and their land use components highly interlinked (Liu et al 2007)developing views about their future is both a major concern in landscape research and a complextask Modelling land use relationships helps to understand underlying drivers of land use changeto create future land use scenarios and assess possible environmental impacts (Lambin and Geist2006 Ravetz 2000)

12 The ldquoidealrdquo urban land use model

A variety of land use change models particularly for urban landscapes already exist ranging fromspecific case studies to generic tools for a variety of urban regions These models differ largely interms of their structure their representation of both space and human decisions and their method-ological implementation Compared to land use change models in open landscapes urban areas areshaped particularly by human activities societal processes and humanndashnature interactions (Coucle-lis 1997) In addition to implemented simulation models a number of articles and book chapterselaborate on the ldquoidealrdquo integrated model theoretically necessary causal feedback loops etc Theseldquoidealrdquo models shall serve as analytical frameworks to better understand the systems under studyOften authors use frameworks like the DPSIR-framework (drivers pressures state impact re-sponses) of the European Environment Agency (EEA) to conceptualise these conceptual modelsAccording to Verburg ldquothe main drawback of using these analytical frameworks is the assumptionof one-directional processes between driving factors and impactsrdquo (Verburg 2006 p 1173) becausein reality it is difficult to differentiate between impacts and drivers in a system Burgi et al (2004)distinguish five major types of driving forces socioeconomic political technological natural andcultural Furthermore they differentiate between primary secondary and tertiary driving forcesas well as between intrinsic and extrinsic driving forces (Burgi et al 2004) In their introductionto urban simulation Waddell and Ulfarsson (2004) sketched urban markets and agents choicesand interactions in an ldquoidealrdquo urban land use model Timmermans (2006) criticizes that presenturban models focus on functional chains like the following demand causes allocation across spacewhich in turn causes traffic flows based upon which a transportation model calculates travel times

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 5

1 Introduction

11 Urbanisation of landscapes

Urbanisation is one of the most complex and dynamic processes of landscape change Althoughonly about 4 of the worldrsquos land area is urbanised and densly populated (Ramankutty et al2006) we claim ldquothe millennium of the citiesrdquo since more than half of the currently 66 billionworld population is living in urban areas (United Nations 2008 2009 PRB 2007 EEA 2006Kasanko et al 2006) Projections for the future show that urbanisation ndash in terms of an increasingshare of population living in urban areas ndash is very likely to continue (Batty et al 2003 EEA2006 Lutz et al 2001) Urbanisation is not only a societal problem but also an environmentalone because it contradicts a normative ideal of ldquoa natural or un-spoiled landscaperdquo in spatialplanning (Nuissl et al 2008) In a multitude of studies it has been shown that land consumption isusually detrimental to the environment in different regards (eg Johnson 2001 Antrop 2004) Itsimpact reduces the ability of landscapes to fulfil human requirements and thus impairs ecosystemservices and landscape functions in various ways (de Groot et al 2002 Millennium EcosystemAssessment 2005 Curran and de Sherbinin 2004) Individual ecosystem services and quality oflife aspects that are affected by urbanisation include the production of food the regulation ofenergy and matter flows water supply the provision of biodiversity and of health and recreationand the supply of green space and natural aesthetic values (Alberti 1999) Suburbanisation andurban sprawl were the dominating land consumption processes in North America and Europeafter WW II (Batty 2008) Recently high growth rates in developing countries have led toenormous environmental loads as discussed above (Heinrichs and Kabisch 2006) As urban systemsare very densely populated and their land use components highly interlinked (Liu et al 2007)developing views about their future is both a major concern in landscape research and a complextask Modelling land use relationships helps to understand underlying drivers of land use changeto create future land use scenarios and assess possible environmental impacts (Lambin and Geist2006 Ravetz 2000)

12 The ldquoidealrdquo urban land use model

A variety of land use change models particularly for urban landscapes already exist ranging fromspecific case studies to generic tools for a variety of urban regions These models differ largely interms of their structure their representation of both space and human decisions and their method-ological implementation Compared to land use change models in open landscapes urban areas areshaped particularly by human activities societal processes and humanndashnature interactions (Coucle-lis 1997) In addition to implemented simulation models a number of articles and book chapterselaborate on the ldquoidealrdquo integrated model theoretically necessary causal feedback loops etc Theseldquoidealrdquo models shall serve as analytical frameworks to better understand the systems under studyOften authors use frameworks like the DPSIR-framework (drivers pressures state impact re-sponses) of the European Environment Agency (EEA) to conceptualise these conceptual modelsAccording to Verburg ldquothe main drawback of using these analytical frameworks is the assumptionof one-directional processes between driving factors and impactsrdquo (Verburg 2006 p 1173) becausein reality it is difficult to differentiate between impacts and drivers in a system Burgi et al (2004)distinguish five major types of driving forces socioeconomic political technological natural andcultural Furthermore they differentiate between primary secondary and tertiary driving forcesas well as between intrinsic and extrinsic driving forces (Burgi et al 2004) In their introductionto urban simulation Waddell and Ulfarsson (2004) sketched urban markets and agents choicesand interactions in an ldquoidealrdquo urban land use model Timmermans (2006) criticizes that presenturban models focus on functional chains like the following demand causes allocation across spacewhich in turn causes traffic flows based upon which a transportation model calculates travel times

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

6 Dagmar Haase and Nina Schwarz

which in turn explain residential choice Timmermans votes to include other aspects of integrationin urban land use models such as task allocation within households residential choice job choicevehicle ownership scheduling of activities competition and agglomeration of land uses and actorsco-evolutionary development of demographics employment sectors land use and activity profilesand a more thorough treatment of varying time horizons including anticipatory and reactive be-haviour According to Miller et al (2004) an integrated urban systems model with a focus ontransport should include socio-demographic components (evolution of population) demographics(demographic change and migration into and out of a region) decision-making (location choices ofhouseholds and firms) economic variables (labour market importexport of goods and services)transportation (activity and travel patterns of population goods and services depending upon ur-ban structure and economic interchanges performance of road and transit systems) and respectiveeffects on land use (evolution of the built environment) and environment (atmospheric emissionsgenerated by transportation and industry Miller et al 2004) Moreover Hunt et al (2005) statedeleven modelling axioms for such an ldquoideal urban land use modelrdquo

bull Representation of an urban system should focus on those elements that interact with thetransportation system

bull An urban system consists of physical elements actors and processes

bull A transportation system is multimodal and involves both people and goods

bull Markets are the basic organising principle of an urban system

bull Flows of people goods information and money arise out of demand

bull Urban areas do not reach an equilibrium

bull System time must be explicitly dealt with

bull Feedback between short-term and long-term processes has to be integrated (eg travel andinfrastructure)

bull Some factors may be treated as exogenous for modelling purposes

bull Some activities arise in response to external demand

bull A very detailed level of representation for actors and processes is necessary

13 Existing reviews on urban land use models

A variety of reviews including urban land use models already exist Agarwal et al (2002) aswell as Schaldach and Priess (2008) review integrated land use models in general also includingmodels that deal with non-urban land uses such as forestry pasture and agriculture Axhausen(2006) specialises in models on transportation demand and traffic flows Beckmann (2006) andIacono et al (2008) focus on interactions between urban land use and transportation The authorspredominantly discuss modelling approaches and does not give details regarding single modelsSimilar to this Berling-Wolff and Wu (2004) provide an historical overview of modelling approachesand do not discuss single models The US EPA (2000) focuses on models of urban growth andsprawl but mainly includes US American approaches and ndash because of its publication date ndashdoes not include recently published models Geurs and van Wee (2004) and Hunt et al (2005)focus on models which emphasize the interaction between urban land use and the transportationsystem Furthermore Timmermans (2006) gives a historical overview and describes a large numberof models but does not give a comparative description of presently developed models With his

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 7

review on modelling the urban ecosystem Alberti (2008) puts less emphasis on urban land usechange but rather focues on the environmental impacts and human-induced environmental stress ofthe urban system The review utilises a range of evaluation criteria of which feedback mechanismsmultiple actors and the inclusion of uncertainy are seen as the most challenging (Alberti 2008)Finally Verburg et al (2004) sketch a few exemplary models but their focus lies on discussinggeneral modelling approaches and not on single causal feedbacks

14 The purpose of this review

Set against the background summarised in Section 13 this review analyses economic models sys-tem dynamics approaches cellular automata and agent-based models developed for urban systemsby systematically addressing a range of critieria such as the conceptual approach model compo-nents and included variables In doing so it aims at giving an overview on the respective modelstructures The main purpose of the review is to derive ideas for causal relationships within landuse change in urban systems with a special emphasis on integrating social and natural science di-mensions The innovative aspect of this review compared to existing reviews is the aim to explicitlyanalyse causalities and feedbacks in urban land use changes

As Verburg (2006) points out an integration of social and biophysical systems could be en-hanced by including feedback mechanisms in land use models eg the feedback between drivingfactors and effects of land use change (here understood as impacts) the feedback between localand regional processes and the feedback between agents and spatial units (Verburg 2006) ldquoLesscommon in land use modelling is the simulation of feedbacks between impacts on socio-economicand environmental conditions and the driving factors of land use changerdquo (Verburg 2006 p 1173)Therefore the review presented here will include a glance at those feedbacks Since urban landuse models deal with spatial entities ndash that is among others the landscape itself ndash an importantaspect of selecting modelling approaches for the review is spatial explicitness in terms of landscapeproperty In addition urban landscapes are highly complex as highlighted in several paragraphsof the introduction part of this paper therefore one should focus on comprehensive models thatinclude different relationships influences and dependencies along with their spatial representationThe paper is organised as follows Section 2 sets up a set of evaluation criteria for conducting themodel review which follows in Section 3 Section 4 especially focuses on causalities and feedbackloops of land use change before coming to the paperrsquos conclusions (Section 5)

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

8 Dagmar Haase and Nina Schwarz

2 Evaluating urban land use simulation models

Compared to natural or agricultural landscapes urban systems are strongly influenced by both thesocial and the natural dimension As mentioned in the introduction urban landscapes are coupledwith humanndashnature systems (Liu et al 2007) with many interlinkages between the human spherendash first and foremost demography and economy ndash land use and the environment Figure 1 providesa very general but comprehensive overview on the major components of an urban landscapethe major driving force for change is the human sphere which creates pressures on the state ofthe land use which again will have effects on the environment its natural resources and ecosystems

Figure 1 Main components (human sphere land use natural resources) and relationships (1 ndash 4) whichdescribe humanndashnature interactions in urban regions (1) Impact of human sphere on land use (2) feedbackof land use on human sphere (3) impact of land use on environment (including ecosystems) and (4) feedbackof environment on human sphere

The human sphere characterises the socio-economic system of cities it comprises variablessuch as population (development) households spatial planning and governance the real estatemarket commercial activities and infrastructure including transportation Specifically the humansphere includes human decision making and actions upon land use The land use componentitself comprehends all types of typical urban land uses such as residential industrial commercialtransport and recreation The third component contains natural resources such as ecosystemsbiodiversity soil functions and water resouces (cf again Figure 1) We set up these feedback loopsbetween the three dimensionscomponents of the urban system discussed above (1) the impactof the human sphere on land use (2) the feedback (= reverse to the impact function) of land useon the human sphere (3) the impact of land use on the environment and (4) another feedbackof the environment on the human sphere All relationships are labelled in Figure 1 respectivelyFurthermore a short Section (44) deals with the scale-specific causal feedbacks between localand regional scale insofar as they are covered by the models investigated The evaluation of thefeedback loops includes (1) the identification of a respective formal representation of the respective

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 9

causalities in the model and (2) whether or not they have an impact on other model componentsagainvice versa In order to structure the review and to give brief overviews of the models underreview we summarised the findings of the analysis of each of the models in Table 1 which providescomprehensive information about the main purpose and major components classified according toFigure 1

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

10 Dagmar Haase and Nina Schwarz

3 Models under review

With respect to the model evaluation criteria mentioned in previous reviews and for the ldquoideal ur-ban modelrdquo (Sections 12 and 13) we solely focus on causalities and feedback loops in the modelsunder review as we believe that alongside a good description of model components (human sphereland use environment) representation of the linkages between the components (= impacts andfeedback loops) make up the comprehensiveness and the explanatory strength of the models Themodels included in this review were selected in order to represent the most influential streams ofurban land use change modelling First the review includes models well known within the com-munity such as those which are discussed in the related literature on urban land use change egby being referenced in other reviews Second system and land use approaches which are not dis-cussed at great length in the literature were included because system dynamics as a method forcesmodellers to think in a systemic way and easily allows for the inclusion of feedback mechanismsFor system-oriented causality-driven models on at least one dimension of urban land-use changea search on the ISI Web of Science was performed This procedure led to a total of 19 modelswhich were also included in this review These models are listed in the form of a comprehensiveoverview in Table 1 Details are given in the Annex 7

Roughly four different modelling approaches can be distinguished Two of the models underreview belong to the class of spatial economicseconometric models (SE 1 and SE 2 Nijkamp et al1993 Mankiw and Weil 1989) These models mainly look at demography and household-drivendemand-supply relations in urban regions such as housing market developments Seven modelsincluded in this review (SD 1 to SD 7 Forrester 1969 Haghani et al 2003ab Eppink et al2004 Sanders and Sanders 2004 Onsted 2002 Eskinasi and Rouwette 2004 Raux 2003) aresystem dynamics or causality-driven models (Table 1) System dynamics is an approach whichmodels complex systems using stocks and flows and by explicitly including feedback loops inthe model (Sterman 2000) System dynamics models are ndash in their standard application ndash notspatially explicit Rather the structure of combining stocks flows and feedback mechanismsleads to a set of differential equations The outcome of these equations can be simulated givenvalues for parameters and initial conditions The classical approach to modelling urban systemsusing system dynamics is Forresterrsquos book on ldquoUrban Dynamicsrdquo (Forrester 1969) He linked thethree subsystems ldquobusinessrdquo ldquohousingrdquo and ldquopopulationrdquo to describe and model urban systems ingeneral subsequently differentiating each of the three subsystems in very detailed sub-models Fivemodels included in this review (CA 1 to CA 5 Verburg and Overmars 2007 Landis and Zhang1998ab Landis et al 1998 Engelen et al 2007 Dietzel and Clarke 2007) use cellular automataas the main modelling technique (Table 1) A cellular automaton consists of an n-dimensionalgrid of cells Each cell has a finite number of states Cells change their state simultaneouslyaccording to the same rules coded in the model and the state of a cell in time t solely dependson the state of neighbouring cells in tminus1 (cf Clarke et al 1997 Landis and Zhang 1998ab Silvaand Clarke 2002) Land use change models use cellular automata with 2-dimensional grids whichrepresent the majority of land use Each cell symbolises a patch of land and states of cells arethe land use options Five models in this review (ABM 1 to ABM 5 Strauch et al 2003 Salviniand Miller 2005 Ettema et al 2007 Loibl et al 2007 Waddell et al 2003) use agent-basedapproaches as the main modelling technique (Table 1) Agent-based models consist of autonomousindividuals (agents) who perceive their environment and interact with one another (Parker et al2003) Applications of agent-based modelling in land use change are usually spatially explicit andagents represent for example households relocating their homes or individuals using transportsystems but also governmental and other institutional bodies

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 11

Table 1 Overview of main purposes and components (according to Figure 1) investigated in reviewedmodels

Model Main purpose Components Reference

Spatial Economics Econometric models

SE 1 Modelling household life cyclesand their impact on residential re-location behaviour and the urbanhousing market for a Europeancapital city

Human sphere (population migra-tion household transportationhousing market prices dwellingsvacancies)

Nijkamp et al(1993)

SE 2 Simulation of demographicchanges (baby boom and babybust) and its influences on thehousing market in the US

Human sphere (population migra-tion household housing marketprices dwellings vacancies)

Mankiw andWeil (1989)

System Dynamics

SD 1 Modelling urban system in gen-eral explicitly including ldquourbandeclinerdquo Examples focus on aspecific topic eg rapid popula-tion growth demolition et ceteraand therefore need specific models

Human sphere (business housingpopulation)

Forrester(1969) Alfeld(1995)

SD 2 Integrated land-use and trans-portation model for estimatingscenarios regarding transport poli-cies

Human sphere (population mi-gration household job growthemployment and commercial landdevelopment housing develop-ment travel demand congestion)

Haghani et al(2003ab)

SD 3 Assessing the impact of urbansprawl on wetland biodiversityand social welfare

Human sphere (population)Land use (agricultural land wet-lands)Environment (wetlands natureprotection)

Eppink et al(2004)

SD 4 Redefining the model of urbandynamics by Forrester (1969) in-cluding 1 spatial dimension (16squares) and 2 disaggregationdifferent types of housing indus-try and people in zones

Human sphere (population hous-ing availability houses land avail-ability business structures andjob availability labour market andhousing market)

Sanders andSanders (2004)

SD 5 Simulation model to provide sce-narios for future land use in SantaBarbara eg with restrictions tourban growth

Human sphere (housing popula-tion business)Land useQuality of life

Onsted (2002)

SD 6 Assessing the impact of futurepolicy interventions on the socialhousing market (specific rate ofbuilding new dwellings)

Human sphere (commercial hous-ing stock social housing stockwaiting families supply of avail-able social houses migration de-molition construction)

Eskinasi andRouwette(2004)

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

12 Dagmar Haase and Nina Schwarz

Table 1 ndash Continued

Model Main purpose Components Reference

SD 7 Simulating medium- and long-termeffects of urban transportationpolicies with reference to sustain-able travel

Human sphere (urbanisation in-ternal travel demand car own-ership external travel demandtransportation socio-economicevaluation)Environment (environmental ap-praisals)

Raux (2003)

Cellular Automata

CA 1 Tool for understanding land-usepatterns possible future scenariosfor given demand

Human sphere (demand rules)Land use (suitability rules)

Verburg andOvermars(2007)

CA 2 Simulating urban growth scenar-ios for future development

Human sphere (population house-hold jobs employment)Land use (single-family residentialmulti-family residential commer-cial industrial transportationpublic)Environment (undeveloped land)

Landisand Zhang(1998ab)

CA 3 Development of policy scenarios ofurban growth impact on habitatchangebiodiversity

Human sphere (urban growthpolicy simulation and evaluation)Environment (habitat change andhabitat fragmentation)

Landis et al(1998)

CA 4 Monitoring developments of urbanareas and identifyng trends at theEuropean level focus is on growthscenarios

Human sphere (population econ-omy planning accessibility viatransportation network)Land use (land use functions)

Engelen et al(2007)

CA 5 Modelling urban growth scenar-ios for future development of anurban region

Land use (urban or non urbanroads different land use types)Environment (topography)

Silva andClarke (2002)Dietzel andClarke (2007)

Agent-Based Models

ABM 1 Dynamic simulation model witha focus on urban traffic flowsincluding activity behaviourchanges in land use and effectson environment

Human sphere (activity patternsand travel demand traffic flowsgoods transport accessibility oflocations location decisions ofhouseholds firms developers)Land use (moving households lo-cation of firms investment of de-velopers new industrial area)Environment (clean air trafficnoise)

Strauchet al (2003)Moeckel et al(2006)

ABM 2 Evolution of an entire urban re-gion with emphasis on transporta-tion

Human sphere (location choice ac-tivity schedule activity patternsautomobile ownership travel de-mand)Land use (land developmenttransportation network)

Salvini andMiller (2005)Miller et al(2004)

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 13

Table 1 ndash Continued

Model Main purpose Components Reference

ABM 3 Predicting urbanisation with be-havioural agents

Human sphere (demographicchange decisions of individuals)

Ettema et al(2007)

ABM 4 Development of built-up areain peri-urban region driven byhouseholds and entrepreneurs ur-ban growth with different growthrates

Human sphere (households jobsnumbers of people householdsand workplaces at the start of theyear average travel time to dis-trict centres and capital city)Land use (urban land open spaceforest area)

Loibl et al(2007)

ABM 5 Link between transport and landuse impact of different planningstrategies

Human sphere (population house-holds employment travel demandaccessibility mobility real estateland price)Land use

Waddell(2006) Wad-dell et al(2003)

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

14 Dagmar Haase and Nina Schwarz

4 Representation of urban landscapes

One of the major aspects which urban land use models have to represent are causalities andfeedbacks related to humanndashnature interactions The main components representing an urbansystem according to the models under review are summarised in Tables 1 and 2 Spatial Economicmodels are labelled SE Cellular Automata CA System Dynamics Models SD and Agent-Basedmodels ABM

Table 2 Main components of urban systems ndash do the models under review include them

Human sphere Land use Environment

(Spatial) Economic models

SE 1 x xSE 2 x x

System dynamics

SD 1 x x xSD 2 x xSD 3 x xSD 4 x xSD 5 x xSD 6 x xSD 7 x

Cellular automata

CA 1 xCA 2 x xCA 3 x x xCA 4 x xCA 5 x

Agent-based models

ABM 1 x xABM 2 x x xABM 3 x xABM 4 x xABM 5 x x

Structural relationships between model components and variables are found to be very differentin the models (Figures 2 and 3) This is due to the fact that levels of rules for land use change varylargely depending on the modelling technique used ie (spatial) economics system dynamicscellular automata or agent behaviour (Table 2)

The first model group (spatial) economic or econometric models sets up a formalised relation-ship between population and market in our case these compounds are the housing market andresidential land use Spatial economics models can be dynamic (when model parameters are treatedendogeneously) or quasi dynamic (if model parameters are fixed or an exogeneous input during themodel runtime) Generally such models define a demand based on a populationhouseholdcohortetc number but only a limited feedback is generated from the net supply to the original driver(in our case population) Cellular automata derive probabilities of land use change for a cer-tain cell out of historical land use data (Engelen et al 2007 Barredo et al 2003) or by usingtry-and-error ldquocalibrationsrdquo (Hansen 2007) Therefore they do not explicitly deal with causalrelationships between urban drivers and land use states Driving forces of the human sphere such

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 15

as population dynamics residential mobility or price elasticise of the real estate market can beincluded as scenario assumptions in some of the models in order to define the magnitude of urbansprawl (eg CA 2 CA 4) Nevertheless the decision about which cells change their land use inwhich way is based upon historical land use change patterns In contrast landscape propertieslike topography hydrography or morphology are reflected in most of the cellular models (CA 1CA 3 ndashCA 5 Table 2) Using a different approach agent-based models include individual and in-stitutional actors to explicitly simulate processes of land use conversion The main actors in thesemodels are individuals or households which choose their residential location according to theirpreferences local industries and businesses which choose their location and employ local peopleand institutions which steer land use development by planning permitting or restricting land usechange et cetera Therefore these models explicitly name the decision-making processes relevantfor urban land use changes (ABM 1 ndashABM 5) System dynamics models lie between these twoldquoextremesrdquo They include the processes but in an aggregate way without incorporating singleactors and their individual goals (Table 2)

In the following the processes captured in the simulation models are analysed with respect tothe feedbacks mentioned in Section 2

41 Spidergrams

For comparison purposes we set up an assessment matrix in which the degree of fulfilment of thefour relationships (cf again Figure 1) is assigned to each of the models under review We used ametric scale from 0 to 2 If the criterion is fulfilled then the ldquomarkrdquo 2 is given if only parts of thecriterion are fulfilled ndash eg the processes implemented by rudimentary or very simple ndash the ldquomarkrdquo1 is given and if the criterion is not at all fulfilled or not included in the model the ldquomarkrdquo 0 isgiven The results of the model assessment are given in forms of simple multicriteria spidergramswhich compare the three types of models (SE SD CA and ABM Figure 2) for all criteria and ina second range of graphs all models for each single criterion (Figure 3)

42 Relationships between human sphere and land use

Most of the models under review represent the impact of human sphere on land use Table 1provides an overview of the model components Except for three model approaches each modelcovers population dynamics and housing or built-up land which belong to the major variableseither for human sphere or land use The spidergram in Figure 2 clearly shows that causal rela-tionships between human drivers are better captured than the reverse feedback from urban landdevelopment to the human drivers Agent-based approaches mainly cover both loops since landuse variables belong to the neighbourhood of the agents and thus directly influence decision mak-ing In comparison spatial economics and system dynamics models comprehensively cover loopsof type 1 ldquohuman sphere to land userdquo but mostly neglect effects of changing urban land use onpopulation dynamics or economic development Cellular automata do include some feedbacks fromthe effects of land use changes on the human sphere

43 Impact of land use on environment

Only very few simulation models close the loop between driving forces and environmental im-pacts Cellular automata perform better in capturing the effects of relatively simple rule-based orneighbourhood-statistic driven land use changes on the environment Since they are often spatiallyexplicit landscapes can be more easily represented (cf again Figure 2) For example in CA 3the impact of urbanisation on biodiversity is assessed but no feedback to driving forces is takeninto account In SD 7 the impact of transport on the environment is integrated but it is not

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

16 Dagmar Haase and Nina Schwarz

0

1

2

SE_1

SE_2

SD_1

SD_2

SD_3SD_4

SD_5

SD_6

SD_7

Loop 1 Loop 2

Loop 3 Loop 4

0

1

2

CA_1

CA_2

CA_3CA_4

CA_5

Loop 1 Loop 2

Loop 3 Loop 4

0

1

2

ABM_1

ABM_2

ABM_3ABM_4

ABM_5

Loop 1 Loop 2

Loop 3 Loop 4

Figure 2 Spidergrams showing how far the reviewed models (according to their model type) incorporatethe four relationships set up for model evaluation

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 17

Loop 1 Human sphere to land use

0

1

2

SE_1

SE_2

SD_1

SD_2

SD_3

SD_4

SD_5

SD_6

SD_7

CA_1CA_2

CA_3

CA_4

CA_5

ABM_1

ABM_2

ABM_3

ABM_4

ABM_5

Loop 2 Land use to human sphere

0

1

2

SE_1

SE_2

SD_1

SD_2

SD_3

SD_4

SD_5

SD_6

SD_7

CA_1CA_2

CA_3

CA_4

CA_5

ABM_1

ABM_2

ABM_3

ABM_4

ABM_5

Loop 3 Land use to environment

0

1

2

SE_1

SE_2

SD_1

SD_2

SD_3

SD_4

SD_5

SD_6

SD_7

CA_1CA_2

CA_3

CA_4

CA_5

ABM_1

ABM_2

ABM_3

ABM_4

ABM_5

Loop 4 Environment to human sphere

0

1

2

SE_1

SE_2

SD_1

SD_2

SD_3

SD_4

SD_5

SD_6

SD_7

CA_1CA_2

CA_3

CA_4

CA_5

ABM_1

ABM_2

ABM_3

ABM_4

ABM_5

Figure 3 Spidergrams showing to which extent all reviewed models include the four loops

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

18 Dagmar Haase and Nina Schwarz

clear from the available literature if there is a feedback to driving forces (travel and transportationflows) The two economics models under review (SE 1 and SE 2) lack spatial explicitness to beable to capture a more comprehensive land use relation or feedback

44 Feedback from environment to human sphere

Feedbacks from environmental impacts back to the driving forces that cause urban land use changeare mostly realised through changing attractiveness of grid cells or regions for household residentiallocation choices Those were found in ABM 4 (open space forest area) ABM 1 (traffic noise airquality) CA 4 (quality and availability of space for activities) and SD 5 (traffic volume producesair pollution and thus affects human quality of life) In SD 3 the decrease of wetland area (andits negative impact on biodiversity) directly influences decisions to buy land for nature protectioninstead of further urbanisation These relationships are the only ones that close the loop fromhouseholdsindividuals as drivers of land use change to environmental impacts and back to theoriginal decision algorithm

45 Feedbacks between local and regional scale

Feedbacks between the local and regional scale can be realised in a variety of ways first migrationof population within single districts can have an influence on the attractiveness of the districtsand therefore influence the housing market in the region which in turn affects migration Secondplanning and governance on the regional scale can influence local land use changes which in turncan impact regional planning In several of the models the housing market (or price development)is captured implicitly or explicitly For example in the spatial economics models SE 1 and SE 2as well as in the system dynamics models SD 2 and SD 4 the housing market and housing devel-opment are explicitly included In the two former cases in the form of real case study examples(Amsterdam and the US) while in SD 2 an artificial market is created between expansionistsand conservationists who want to buy open land ndash either in order to turn it into urban area orto conserve it In cellular automata prices for housing are not explicitly included Probabili-ties for land use change can be regarded as bids for (re-)development (CA 2) In some of theagent-based models real estate markets are already included or are planned to be included (egABM 1 ABM 3 ABM 5) In these models developers are agents who can influence the marketand therefore also the prices Governmental planning processes are never explicitly represented ina way that governmental agencies are actors within the model In some models planning decisionsare integrated as a part of the scenario configuration eg by restricting or promoting possibleevolution paths for certain grid cells (eg MOLAND) In others construction and demolition areexogeneous variables (Nijkamp et al 1993) But in these cases planning decisions or housingmarket trends are not changed during the simulation so that no feedbacks are established

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 19

5 Conclusions

The main purpose of this review was to analyse causalities and feedback loops in current urbanland use change models Therefore we analysed 19 simulation models stemming from four differentsimulation methodologies spatial economics system dynamics cellular automata and agent-basedmodelling The main conclusion of this review is that there is a range of comprehensive urbanland use change models but no unique approach to represent urban landscapes and humanndashnatureinteractions Each author or working group has its own view and focuses on other parts of theurban system and the relationships within that system Thus the landscape aspect is of minorimportance Most of the approaches bear the potential to model local and regional urban processesas they provide a multitude of components and variables However currently only a few modelsintegrate direct or indirect feedback loops from environmental and landscape-related impacts ofurban land use change on environment to the respective driving forces in the human sphere ofthe systems We see the reason for this in the gap between social science methods and findingsand computational models (cf Geist and Lambin 2004 2002) The former comprehensively coverbehavioural heuristics on decision making but are often qualitative in nature The latter needquantitative (sometimes spatially explicit) input data or at least simple rules to be coded andthus incorporated into the models To bring both approaches together and to better incorporatequalitative social science data into quantitative models is still one of the major challenges of urbanland use and landscape modelling This is a challenge not only for modellers since empirical datafor formulating a resilient feedback loop resulting from environmental impacts on human qualityof life and decision making is rarely available (Haase and Haase 2008) As urban systems areopen systems which do not depend on local or regional natural resources and ecosystem servicesneither individual nor policy decisions strongly depend on the availability and state of nature ofthe surroundings (cf Haase and Nuissl 2007) This makes it more difficult to elicit and formaliseresilient feedbacks from the environment or landscape back to the driver Another challenge is toexpress urban land use relationships and in particular the aforementioned decision making in aspatially explicit way as most of the CA models under review do Finally relationships betweenthe local and regional scale are realised only with respect to housing markets as single choiceson the local scale are able to influence regional markets and vice versa None of the models dealswith all possible linkages between ldquothe built-up urbanrdquo and ldquothe ruralrdquo landscape within an urbanregion although CA models such as MOLAND cover both types of land use at least in termsof land use types Current ldquohotspotsrdquo of the worldwide agri-environmental discussion such asbiofuels and organic farming should also be partially incorporated into urban models Here wesee another way to introduce more landscape aspects into urban land use modelling

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

20 Dagmar Haase and Nina Schwarz

6 Acknowledgements

This work is based on a deliverables for the PLUREL Integrated Project (Peri-urban Land UseRelationships) funded by the European Commission Directorate-General for Research under the6th Framework Programme (project reference 36921) This paper reflects the authorsrsquo views andnot those of the European Community Neither the European Community nor any member of thePLUREL Consortium is liable for any use of the information in this paper The authors thankKlaus Steinocher Nick Green and anonymous reviewers for comments on earlier versions of thismanuscript

7 Annex

Within the following tables (describing the models in alphabetical order) empty cells indicatethat no information was found in the literature on this issue ldquondashrdquo in a cell means that this issue isnot applicable to the model in question

Field ldquoDuration of model runrdquo

bull C Calibration to fit model parameters

bull S Scenarios for projections of future trends

bull V Validation using independent data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 21

Table 3 Household life cycle model for residential relocation behaviour [SE 1]

Name of model Household life cycle model for residential relocation behaviour

Sources Nijkamp et al (1993)

Technical data

Coveredarea physicalboundaries

Case studyGreater Amster-dam Area

Extent of area 350 square miles About 800000peopleApplication area

Spatial units 20 zones Size or grain ofgridszones

ndash

Time horizon Time step 1 year Duration ofmodel run

1971 ndash 1984

Modelling ap-proach

Simulationtechnique

Spatial economics Qualitative orquantitative

Quantitative

Contents

Main purpose Modelling household life cycles and their impact on residential relocation be-haviour and the urban housing market for a European capital city

Main variableswith relationships

(1) households (2) migration (3) occupancy (4) housing demand (5) dwellingsupply in zones and dwelling types (6) allocation of households

Domain Not explicitly Temporalrange

ndash

Typology(classes) ofagents

Allocation ofhousehold

rarr if yes whattypes

Households single2-person house-hold 3-personhousehold 4+person householdnon-household

Human decisionmaking

Decision algo-rithm

Rational choicemaximum utility

Input into de-cision

Population andhousehold data

Goals Authorsrsquo opin-ion

Successful runs validation and scenarios

Model developmentprocess

Concept Given Quantificationof relationships

Empirical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

22 Dagmar Haase and Nina Schwarz

Table 4 Simulation of demographic changes and the housing market [SE 2]

Name of model Simulation of demographic changes and the housing market

Sources Mankiw and Weil (1989)

Technical data

Coveredarea physicalboundaries

US cities (census) Extent of area ndash 203190 people 74565 house-holdsApplication area

Spatial units US cities (census) Size or grain ofgridszones

ndash

Time horizon Time step 1 year Duration ofmodel run

1970 ndash 2007 or 2020

Modelling ap-proach

Simulationtechnique

Spatial economics Qualitative orquantitative

Quantitative

Contents

Main purpose Simulation of demographic changes (baby boom and baby bust) and its influ-ences on the housing market in the US

Main variableswith relationships

(1) population (2) households (3) housing market (demand prices) (4) econ-omy (GNP)

Domain Not explicitly Temporalrange

ndash

Typology(classes) ofagents

Allocation ofhousehold

rarr if yes whattypes

Dummy household

Human decisionmaking

Decision algo-rithm

Rational choicemaximum utility

Input into de-cision

Census data

Goals Authorsrsquo opin-ion

Successful runs validation and scenarios

Model developmentprocess

Concept Given Quantificationof relationships

Empirical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 23

Table 5 A System Dynamics Approach to Land Use Transportation System Performance Modeling[SD 2]

Name of model A System Dynamics Approach to Land Use Transportation System Perfor-mance Modeling

Sources Haghani et al (2003ab)

Technical data

Coveredarea physicalboundaries

Varies with appli-cation area Casestudy MontgomeryCounty

Extent of area ndash About 800000people

Application area

Spatial units US cities (census) Size or grain ofgridszones

ndash

Time horizon Time step 1 year Duration ofmodel run

C 1970 ndash 1980V 1980 ndash 1990

Modelling ap-proach

Simulationtechnique

Spatial economics Qualitative orquantitative

Quantitative

Contents

Main purpose Integrated land-use and transportation model for estimating scenarios regard-ing transport policies

Main variableswith relationships

Seven sub-models (1) population (2) migration (3) household (4) jobgrowth employment and commercial land development (5) housing devel-opment (6) travel demand and (7) congestion

Domain Not explicitly Temporalrange

ndash

Typology(classes) ofagents

Cohorts withinpopulation sub-model

rarr if yes whattypes

Persons age 0ndash1718ndash44 45ndash64 65male and female Households singlemarried with chil-dren married with-out children maleor female with chil-dren other

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

First step is achieved successful validation and scenar-ios

Model developmentprocess

Concept Not stated Quantificationof relationships

Empirical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

24 Dagmar Haase and Nina Schwarz

Table 6 CLUE-s (Conversion of Land Use and its Effects) [CA 1]

Name of model CLUE-s (Conversion of Land Use and its Effects)

Sources Verburg and Overmars (2007)

Technical data

Coveredarea physicalboundaries

User-specified Several examplespublished

Extent of area User-specified

Application area

Spatial units CLUE soft-classified data(large pixels withfraction of land-uses)

Size or grain ofgridszones

User-specified CLUE 7 to 32 km CLUE-s20 to 1000 m

Time horizon Time step Iterative processstops when de-mand for land-usemeets allocatedarea

Duration ofmodel run

ndash

Modelling ap-proach

Simulationtechnique

Cellular automata Qualitative orquantitative

Quantitative

Contents

Main purpose Tool for understanding land-use patterns possible future scenarios for givendemand

Main variableswith relationships

Input Pre-defined change in demand for land by different sectors for wholesimulation area rarr CLUE-s assigns new land-uses per gridEach cell most preferred land use based on suitability of location and com-petitive advantage of different land use types (demand) check is land usechange allowed If no next most preferred land use is chosen

Domain Not explicit deci-sion making

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Case-study specific

Model developmentprocess

Concept Not mentioned Quantificationof relationships

User-specified em-pirical analysisexpert knowledgespatial interactionsconversion elastici-ties

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 25

Table 7 CUF-2 (California Urban Futures) [CA 2]

Name of model CUF 2 (California Urban Futures)

Sources Landis and Zhang (1998ab)

Technical data

Coveredarea physicalboundaries

San Francisco BayArea (California)

Extent of area 18 million ha

Application area

Spatial units Grid cells Size or grain ofgridszones

100 times 100 m

Time horizon Time step Econometric10 years Prob-abilities for landuse change onceper simulation

Duration ofmodel run

C 1985 ndash 1995 S

Modelling ap-proach

Simulationtechnique

Cellular automata Qualitative orquantitative

Quantitative

Contents

Main purpose Simulating urban growth scenarios for future development

Main variableswith relationships

Top-down approach future trends of population household jobs rarr areassigned to grid cellsEconometric models predict future population households employment(10 year intervals)LUC-model estimates probabilities for land use change out of historicaldata and simulation engine assigns probabilities to cellsProbability of land use change (multinomial logit models) for a cell fromi to j = f (initial site use site characteristics site accessibility communitycharacteristics policy factors relationships to neighbouring sites) rarr proba-bilities are interpreted as bids for (re-)development rarr population and jobsare assigned to cells by bids7 urban land-use categories undeveloped single-family residential multi-family residential commercial industrial transportation public

Domain Not explicit deci-sion making

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Achieved

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Calibration usingmaps of land usechange

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

26 Dagmar Haase and Nina Schwarz

Table 8 CURBA (California Urban and Biodiversity Analysis)(CA 3]

Name of model CURBA (California Urban and Biodiversity Analysis)

Sources Landis et al (1998)

Technical data

Coveredarea physicalboundaries

San Francisco BayArea (California)

Extent of area See CUF-2

Application area

Spatial units Grid cells Size or grain ofgridszones

100 times 100 m

Time horizon Time step Duration ofmodel run

Modelling ap-proach

Simulationtechnique

Cellular automata Qualitative orquantitative

Quantitative

Contents

Main purpose Development of policy scenarios of urban growth impact on habitatchangebiodiversity

Main variableswith relationships

Two components (1) urban growth model and (2) policy simulation andevaluation model Urban growth model is based upon CUF-2Policy simulation and evaluation several growth scenarios rarr impact onhabitat change and habitat fragmentation

Domain No explicit decisionmaking

Temporalrange

-

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Achieved

Model developmentprocess

Concept See CUF-2 Quantificationof relationships

See CUF-2

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 27

Table 9 ILUMASS (Integrated Land-Use Modelling and Transportation System Simulation) [ABM 1]

Name of model ILUMASS (Integrated Land-Use Modelling and Transportation System Sim-ulation)

Sources Strauch et al (2003) Moeckel et al (2006)

Technical data

Coveredarea physicalboundaries

Dortmund and its25 surroundingmunicipalities

Extent of area About 2000 km2 26 million people

Application area

Spatial units Statistical zones(total 246) andgrid cells

Size or grain ofgridszones

Grid cells100 times 100 m

Time horizon Time step One year Duration ofmodel run

S 2000 ndash 2030

Modelling ap-proach

Simulationtechnique

Coupled simulationsystem includingagent-based simu-lations

Qualitative orquantitative

Quantitative

Contents

Main purpose Dynamic simulation model with a focus on urban traffic flows including ac-tivity behaviour changes in land use and effects on environment

Main variableswith relationships

Five modules (+ integration module) 1 changes in land use 2 activitypatterns and travel demand 3 traffic flows 4 goods transport 5 environ-mental impacts of transportation and land useLand use rarr demand for spatial interaction (work shopping trips etc) rarrtraffic rarr environmental impactsFeedbacks (a) transport rarr accessibility of locations rarr location decisionsof households firms developers (b) environmental factors rarr location deci-sions (eg clean air traffic noise)Land use module moving households location of firms investment of de-velopers new industrial area

Domain Various eg trans-port householdlocation daily ac-tivity plans

Temporalrange

Depending upondomain (dailytravel behaviourvs moving)

Typology(classes) ofagents

Yes rarr if yes whattypes

Not mentioned

Human decisionmaking

Decision algo-rithm

Various (MarkovLogit Monte-Carlo)

Input into de-cision

Depending upondomain feedbacksincluded

Goals Authorsrsquo opin-ion

Time of report work in progress later papers all focuson single modules

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Not mentioned

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

28 Dagmar Haase and Nina Schwarz

Table 10 ILUTE (Integrated Land Use Transportation Environment model) [ABM 2]

Name of model ILUTE (Integrated Land Use Transportation Environment model)

Sources Salvini and Miller (2005) Miller et al (2004)

Technical data

Coveredarea physicalboundaries

Tests for Torontoarea

Extent of area 5 million people

Application area

Spatial units Two versions gridsand buildings

Size or grain ofgridszones

2 parallel ap-proaches Grid30 times 30 m Build-ings as objects

Time horizon Time step Varying with sub-models

Duration ofmodel run

V 1986 ndash 2001 S10 ndash 20 years intofuture

Modelling ap-proach

Simulationtechnique

Agent-based simu-lation

Qualitative orquantitative

Quantitative

Contents

Main purpose Evolution of an entire urban region with emphasis on transport

Main variableswith relationships

Land development rarr location choice rarr activity schedule rarr activity patternsrarr back to land development and all other variables in chain transportationnetwork rarr automobile ownership rarr travel demand rarr network flows rarr backto transportation network and all other variables in chain influences

Domain Activitytravellingscheduling routechoice real es-tate market be-haviour of econ-omy land devel-opment householdownership

Temporalrange

Depends upon do-main Eg typicaltravel day is com-puted once persimulation year peragent type

Typology(classes) ofagents

Yes rarr if yes whattypes

For householdsindividuals firms

Human decisionmaking

Decision algo-rithm

Rule-based re-ducing numberof choices logitmodel for selectingthe ldquobestrdquo option

Input into de-cision

Not mentioned

Goals Authorsrsquo opin-ion

Work in progress

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Empirical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 29

Table 11 Modelling biodiversity and land use [SD 3]

Name of model Modelling biodiversity and land use

Sources Eppink et al (2004)

Technical data

Coveredarea physicalboundaries

No explicit rep-resentation of aspecific area Ur-ban region withsurrounding areaincluding wetlands

Extent of area ndash

Application area

Spatial units No spatial resolu-tion

Size or grain ofgridszones

ndash

Time horizon Time step 1 year Duration ofmodel run

S 100 years

Modelling ap-proach

Simulationtechnique

System dynamics Qualitative orquantitative

Qualitative

Contents

Main purpose Assessing the impact of urban sprawl on wetland biodiversity and social welfare

Main variableswith relationships

Population growth within city rarr higher population density and more needfor agricultural land rarr expansionists attempt to buy surrounding area rarrchange of wetland area to urban area amp more agriculture decrease wetlandbiodiversity rarr conservationistsrsquo valuation of remaining biodiversity increasesrarr conservationists buy wetland area for nature protection

Domain Human decisionmaking is rep-resented withinsystem dynamicsequations

Temporalrange

1 year

Typology(classes) ofagents

Yes rarr if yes whattypes

Expansionists con-servationists (seeabove) and ownersof land

Human decisionmaking

Decision algo-rithm

Land is sold to thehighest bidder

Input into de-cision

Prices offered byconservationistsand expansionists

Goals Authorsrsquo opin-ion

First step for improving relationship between economicdevelopment and biodiversity

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Not mentioned

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

30 Dagmar Haase and Nina Schwarz

Table 12 MOLAND [CA 4]

Name of model MOLAND

Sources Engelen et al (2007)

Technical data

Coveredarea physicalboundaries

Several examplesacross Europe andelsewhere

Extent of area User-specified

Application area

Spatial units global 1 zone regional zonestypically NUTS local grid cells

Size or grain ofgridszones

User-specified

Time horizon Time step annual Duration ofmodel run

C last 40 ndash 50years S user-specified normally30 years

Simulationtechnique

Mainly rule-basedcellular automata

Qualitative orquantitative

Quantitative

Contents

Main purpose To monitor developments of urban areas and identify trends at the Europeanlevel focus is on growth scenarios

Main variableswith relationships

Growth of economy and population (global level) rarr growth in competingregions (regional level) sets boundaries for all cells in a region rarr rules forland use change at the grid level physical suitability institutional suitabil-ity (eg planning documents) accessibility (via transport network) dy-namics at the local level (land use functions attracting or repelling eachother)Feedback from grid level to regional level spatial distribution leads to qual-ity and availability of space for different activities which influences compar-ative attractiveness of a region

Domain No explicit decisionmaking

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Achieved

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Calibration withhistorical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 31

Table 13 PUMA (Predicting Urbanisation with Multi-Agents) [ABM 3]

Name of model PUMA (Predicting Urbanisation with Multi-Agents)

Sources Ettema et al (2007)

Technical data

Coveredarea physicalboundaries

North DutchRanstadt (includ-ing AmsterdamUtrecht Schipholairport)

Extent of area 316 million inhabi-tants

Application area

Spatial units Grid cells (andtravel zones)

Size or grain ofgridszones

500 times 500 m

Time horizon Time step 1 year later upto daily

Duration ofmodel run

S 2000 to approx2050

Modelling ap-proach

Simulationtechnique

Agent-based simu-lation

Qualitative orquantitative

Quantitative

Contents

Main purpose Predicting urbanisation using behavioural agents

Main variableswith relationships

Demographic change rarr decisions of individuals rarr land use change Not yetimplemented developers authorities and firmsinstitutions (so far exogenous)[impact of householdrsquos decisions on land use not described]

Domain 1 demographicevents (no deci-sions just stochas-tic)2 residentialrelocation3 job changes

Temporalrange

Annual [Daily de-cisions in futurework]

Typology(classes) ofagents

Yes rarr if yes whattypes

Households Num-ber of adults andchildren age ofhousehold head[dwellings areagents as well]

Human decisionmaking

Decision algo-rithm

Rational choicewith utility max-imisation

Input into de-cision

Residential relo-cation character-istics of dwellingcommuting dis-tance socio-demographics Job choice salaryjob type dis-tance to dwellingpersonal prefer-ences

Goals Authorsrsquo opin-ion

Promising approach still work in progress

Model developmentprocess

Concept Empirical data Quantificationof relationships

Empirical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

32 Dagmar Haase and Nina Schwarz

Table 14 Rotterdam urban dynamics [SD 4]

Name of model Rotterdam urban dynamics

Sources Sanders and Sanders (2004)

Technical data

Coveredarea physicalboundaries

Rotterdam Extent of area 100000 acres

Application area

Spatial units 16 grid cells calledldquozonesrdquo

Size or grain ofgridszones

Squares with 3125miles each side

Time horizon Time step Duration ofmodel run

S 250 years

Modelling ap-proach

Simulationtechnique

System dynamics Qualitative orquantitative

Quantitative

Contents

Main purpose Redefining the model of urban dynamics by Forrester (1969) including 1spatial dimension (16 squares) and 2 disaggregation different types of hous-ing industry and people in zones

Main variableswith relationships

Bi-directional causal loops between population housing availability housesland availability business structures and job availability (linked with popu-lation) Two markets labor market and housing market compete for land (no transportation)

Domain No explicit decisionmaking

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Case of Rotterdam only as an example for generic re-sults

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Out of statisticaldata and expertknowledge

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 33

Table 15 SCOPE (South Coast Outlook and Participation Experience) [SD 5]

Name of model SCOPE (South Coast Outlook and Participation Experience)

Sources Onsted (2002)

Technical data

Coveredarea physicalboundaries

South Coast ofSanta BarbaraCounty

Extent of area 137000 acres Approx 200000inhabitantsApplication area

Spatial units No spatial resolu-tion

Size or grain ofgridszones

ndash

Time horizon Time step Duration ofmodel run

V 1960 ndash 2000 S2000 ndash 2040

Modelling ap-proach

Simulationtechnique

System dynamics Qualitative orquantitative

Quantitative

Contents

Main purpose Simulation model to provide scenarios for future land use in Santa Barbaraeg with restrictions to urban growth

Main variableswith relationships

Five sectors housing population business quality of life land use

Domain No explicit decisionmaking

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Achieved but should still become more differentiated

Model developmentprocess

Concept Expert knowledge Quantificationof relationships

Assumptions andstatistical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

34 Dagmar Haase and Nina Schwarz

Table 16 Simulation of polycentric urban growth dynamics through agents [ABM 4]

Name of model Simulation of polycentric urban growth dynamics through agents

Sources Loibl et al (2007)

Technical data

Coveredarea physicalboundaries

Austrian Rhinevalley withmedium-sized cen-tres and rural vil-lages

Extent of area 7330 hectaresbuilt-up area 260000 inhabitants

Application area

Spatial units Grid cells Size or grain ofgridszones

50 times 50 m cells

Time horizon Time step Simulation stopswhen certainhousehold popula-tion and workplacegrowth numbersare achieved

Duration ofmodel run

V 1990 ndash 2000 Suser-specified

Modelling ap-proach

Simulationtechnique

Agent-based simu-lation

Qualitative orquantitative

Quantitative

Contents

Main purpose Development of built-up area in peri-urban region driven by households andentrepreneurs urban growth with different growth rates

Main variableswith relationships

Initialisation increase of household and workplace numbers is defined1 Municipality choice depending on regional attractiveness criteria (num-bers of people households and workplaces in the start of the year averagetravel time to district centres and capital city average share of attractiveland-use classes in the municipality (open space forest area) rarr householdgrowth and workplace growth per municipality rarr transformation of abso-lute values into relative search frequencies rarr agents choose municipality viadiscrete choice2 Local target area search start with random cell choosing most attrac-tive cell3 land use change (new built-up area higher density) rarr influencing localattractivenessDomain Causing the con-

struction of newbuilt-up area orthe densifica-tion of existingarea no movingas lsquoexchangersquo ofdwellings

Temporalrange

Long-term (mov-ing start-up ofcompanies)

Typology(classes) ofagents

Yes rarr if yes whattypes

Four householdtypes (1 2 3 or4 persons) andtwo entrepreneurs(small and large)

Human decisionmaking

Decision algo-rithm

Discrete choice Input into de-cision

Regional and localattractiveness

Goals Authorsrsquo opin-ion

Achieved

Model developmentprocess

Concept Empirical data Quantificationof relationships

Empirical data

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 35

Table 17 SLEUTH (Slope Landuse Exclusion Urban Extend Transportation and Hillshade) [CA 5]

Name of model SLEUTH (Slope Landuse Exclusion Urban Extend Transportation and Hill-shade)

Sources Clarke et al (1997) Silva and Clarke (2002) Dietzel and Clarke (2007)

Technical data

Coveredarea physicalboundaries

Numerous applica-tions mostly US

Extent of area User-specified

Application area

Spatial units Grid cells Size or grain ofgridszones

Input for model8-bit GIF(100 times 100 m cellscan be converted)

Time horizon Time step 1 year Duration ofmodel run

C at least 4 timesteps S User-specified

Modelling ap-proach

Simulationtechnique

Cellular automata Qualitative orquantitative

Quantitative

Contents

Main purpose Modelling urban growth scenarios for future development of an urban region

Main variableswith relationships

Two components (use depends on available data)(1) Urban growth cells have one of two states urban or non urban(2) Urban land use change with different land-use typesFour types of growth behaviour spontaneous diffusive (with new growthcentres) organic (into surroundings) and road-influencedFive main coefficients diffusion breed spread slope and road coefficient(need to be calibrated for each case study)Self modification rules eg concerning the kind of exponential or S-curvegrowth denser road network rarr road gravity factor increases land avail-ability decreases rarr slope resistance factor is decreased (more hilly areas)spread factor increases over time

Domain No explicit decisionmaking

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Achieved

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Calibration usinghistorical maps

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

36 Dagmar Haase and Nina Schwarz

Table 18 Urban dynamics [SD 1]

Name of model Urban dynamics

Sources Forrester (1969) Alfeld (1995)

Technical data

Coveredarea physicalboundaries

Either suburban orcore area (Forrester1969 2) Exam-ples mentioned inAlfeld 1995 Low-ell Boston Con-cord MarlboroughPalm Coast

Extent of area User-specified

Application area

Spatial units No spatial resolu-tion

Size or grain ofgridszones

ndash

Time horizon Time step Duration ofmodel run

S Up to 250 years

Modelling ap-proach

Simulationtechnique

System dynamics Qualitative orquantitative

Quantitative

Contents

Main purpose Modelling urban system in general explicitly including ldquourban declinerdquo Ex-amples focus on a specific topic eg rapid population growth demolition etcetera and therefore need specific models

Main variableswith relationships

Original model by Forrester Three subsystems business housing population

Domain No explicit decisionmaking

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Achieved

Model developmentprocess

Concept Expert knowledge Quantificationof relationships

Statistical dataand own estimation

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 37

Table 19 Urban transformation process in the Haaglanden region [SD 6]

Name of model Simulating the urban transformation process in the Haaglanden region in theNetherlands

Sources Eskinasi and Rouwette (2004)

Technical data

Coveredarea physicalboundaries

The Haaglandenregion includingthe Hague and sur-rounding suburbs

Extent of area

Application area

Spatial units No spatial resolu-tion

Size or grain ofgridszones

ndash

Time horizon Time step Duration ofmodel run

S 1998 ndash 2010

Modelling ap-proach

Simulationtechnique

System dynamics Qualitative orquantitative

Qualitative

Contents

Main purpose Assessing the impact of future policy interventions on the social housing mar-ket (specific rate of building new dwellings)

Main variableswith relationships

Four stocks1 Commercial housing stock2 Social housing stock3 Waiting families4 Supply of available social housesProcesses involved Migration demolition construction

Domain No explicit decisionmaking

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Model is useful for its goal

Validation No (but impact ofprocess on stake-holders is moni-tored)

Plausibilityanalysis

With stakeholders

Model developmentprocess

Concept Participation ofstakeholders nar-rative approach

Quantificationof relationships

Empirical data orexpert guesses

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

38 Dagmar Haase and Nina Schwarz

Table 20 Urban travel system [SD 7]

Name of model A system dynamics model for the urban travel system

Sources Raux (2003)

Technical data

Coveredarea physicalboundaries

Hypothetical city Extent of area ndash

Application area

Spatial units No spatial resolu-tion

Size or grain ofgridszones

ndash

Time horizon Time step Duration ofmodel run

S 20 years into thefuture

Modelling ap-proach

Simulationtechnique

System dynamics Qualitative orquantitative

Quantitative

Contents

Main purpose To simulate medium- and long-term effects of urban transport policies withreference to sustainable travel

Main variableswith relationships

Seven major blocks urbanisation internal travel demand (trips withinsystem) car ownership external travel demand (inflowing outflowing andthrough traffic) transportation (comparing supply and demand) and evalua-tion (socioeconomic and environmental appraisals)

Domain No explicit decisionmaking

Temporalrange

ndash

Typology(classes) ofagents

ndash rarr if yes whattypes

ndash

Human decisionmaking

Decision algo-rithm

ndash Input into de-cision

ndash

Goals Authorsrsquo opin-ion

Work in progress

Model developmentprocess

Concept Expert knowledge Quantificationof relationships

Expert knowledgeand statistical val-ues

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 39

Table 21 UrbanSim [ABM 5]

Name of model UrbanSim

Sources Waddell (2006) Waddell et al (2003)

Technical data

Coveredarea physicalboundaries

Several examplesin the US Europeand Asia

Extent of area User-specified

Application area

Spatial units Initially mix-ture of parcels andzones later grid

Size or grain ofgridszones

User-specified Cell 150 times 150 mregarded as default

Time horizon Time step 1 year Duration ofmodel run

User-specified

Modelling ap-proach

Simulationtechnique

Coupled simulationmodels includingagent-based simu-lations

Qualitative orquantitative

Quantitative

Contents

Main purpose Link between transportation and land use impact of different planning strate-gies

Main variableswith relationships

Exogenous (1) macroeconomics (population employment) and (2) traveldemand (travel conditions) Six models1 Accessibility (output access to workplaces and shops for each cell)2 Transition (output number of new jobs and new households per year)3 Mobility (output number of moving (existing) jobs households)4 Location (output location of new or moving jobs households)5 Real Estate Development (output land use change)6 Land price (output land prices)

Domain Mobility and loca-tion

Temporalrange

Depends on issues

Typology(classes) ofagents

Initially households firms later per-sons jobs

rarr if yes whattypes

User-specified

Human decisionmaking

Decision algo-rithm

Multinomial logitmodel

Input into de-cision

Land-use it-self socio-demographicsdwellings

Goals Authorsrsquo opin-ion

Achieved

Model developmentprocess

Concept Not mentioned Quantificationof relationships

Out of empiricaldata

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

40 Dagmar Haase and Nina Schwarz

References

Agarwal C Green GM Grove JM Evans TP Schweik CM (2002) ldquoA Review andAssessment of Land-Use Change Models Dynamics of Space Time and Human Choicerdquo GenTech Rep NE-297 Newton Square PA (USDA Forest Service Northern Research Station)Related online version (cited on 2 April 2009)httpnrsfsfeduspubsgtrgtr_ne297pdf 13

Alberti M (1999) ldquoUrban Patterns and Environmental Performance What Do We KnowrdquoJournal of Planning Education and Research 19(2) 151ndash163 doi1011770739456X990190020511

Alberti M (2008) ldquoModeling the Urban Ecosystem A Conceptual Frameworkrdquo in Urban EcologyAn International Perspective on the Interaction Bewteen Humans and Nature (Eds) MarzluffJM Shulenberger E Endlicher W Alberti M Bradley G Ryan C Simon U ZumBrun-nen C New York (Springer) doi101007978-0-387-73412-5 41 13

Alfeld LE (1995) ldquoUrban dynamics ndash The first fifty yearsrdquo System Dynamics Review 11(3)199ndash217 doi101002sdr4260110303 1 18

Antrop M (2004) ldquoLandscape change and the urbanization process in Europerdquo Landscape andUrban Planning 67(1ndash4) 9ndash26 doi101016S0169-2046(03)00026-4 11

Axhausen KW (2006) ldquoNeue Modellansatze der Verkehrsnachfragesimulation Entwicklungslin-ien Stand der Forschung Forschungsperspektivenrdquo in Integrierte Mikro-Simulation von Raum-und Verkehrsentwicklung Theorie Konzepte Modelle Praxis (Ed) Beckmann KJ Tagungs-band des 9 Aachener Kolloquium lsquoMobilitat und Stadtrsquo (AMUS 2006) 18 ndash 19 September 2006vol 81 of Schriftenreihe Stadt Region Land pp 149ndash163 Aachen (RWTH) Related online ver-sion (cited on 8 April 2009)httpe-collectionethbibethzchvieweth28951 13

Barredo JI Kasanko M McCormick N Lavalle C (2003) ldquoModelling dynamic spatial pro-cesses Simulation of urban future scenarios through cellular automatardquo Landscape and UrbanPlanning 64(3) 145ndash160 doi101016S0169-2046(02)00218-9 4

Batty M (2008) ldquoThe Size Scale and Shape of Citiesrdquo Science 319 769ndash771 doi101126science1151419 11

Batty M Besussi E Chin N (2003) ldquoTraffic Urban Growth and Suburban Sprawlrdquo CASAWorking Papers Series 70 London (University College London) Related online version (citedon 2 April 2009)httpwwwcasauclacukworking_paperspaper70pdf 11

Beckmann KJ (2006) ldquoMikro-Simulation von Raum- und Verkehrsentwicklung ndash Stand der Kunstund Perspektiven zwischen Forschung Entwicklung und Praxisrdquo in Integrierte Mikro-Simulationvon Raum- und Verkehrsentwicklung Theorie Konzepte Modelle Praxis (Ed) BeckmannKJ Tagungsband des 9 Aachener Kolloquium lsquoMobilitat und Stadtrsquo (AMUS 2006) 18 ndash 19September 2006 vol 81 of Schriftenreihe Stadt Region Land pp 1ndash31 Aachen (RWTH) 13

Berling-Wolff S Wu JG (2004) ldquoModeling urban landscape dynamics A reviewrdquo EcologicalResearch 19(1) 119ndash129 doi101111j1440-1703200300611x 13

Burgi M Hersperger AM Schneeberger N (2004) ldquoDriving forces of landscape change ndashcurrent and new directionsrdquo Landscape Ecology 19(8) 857ndash868 doi101007s10980-004-0245-8 12

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 41

Clarke KC Hoppen S Gaydos L (1997) ldquoA self-modifying cellular automaton model of his-torical urbanization in the San Francisco Bay areardquo Environment and Planning B Planningand Design 24(2) 247ndash261 doi101068b240247 3 17

Couclelis H (1997) ldquoFrom cellular automata to urban models new principles for model develop-ment and implementationrdquo Environment and Planning B Planning and Design 24(2) 165ndash174doi101068b240165 12

Curran SR de Sherbinin A (2004) ldquoCompleting the Picture The Challenges of BringinglsquoConsumptionrsquo into the PopulationndashEnvironment Equationrdquo Population and Environment 26(2) 107ndash131 doi101007s11111-004-0837-x 11

de Groot RS Wilson MA Boumans RMJ (2002) ldquoA typology for the classification de-scription and valuation of ecosystem functions goods and servicesrdquo Ecological Economics 41(3) 393ndash408 doi101016S0921-8009(02)00089-7 11

Dietzel C Clarke KC (2007) ldquoToward Optimal Calibration of the SLEUTH Land Use ChangeModelrdquo Transactions in GIS 11(1) 29ndash45 doi101111j1467-9671200701031x 3 1 17

EEA (2006) ldquoUrban sprawl in Europ The ignored challengerdquo EEA Report 10 Copenhagen(European Environmental Agency) Related online version (cited on 2 April 2009)httpwwweeaeuropaeupublicationseea_report_2006_10 11

Engelen G Lavalle C Barredo JI van der Meulen M White R (2007) ldquoThe MolandModelling Framework for Urban and Regional Land-use Dynamicsrdquo in Modelling Land-UseChange Progress and Applications (Eds) Koomen E Stillwell J Bakema A ScholtenHJ pp 297ndash319 Dordrecht (Springer) doi101007978-1-4020-5648-2 3 1 4 12

Eppink FV van den Bergh JCJM Rietveld P (2004) ldquoModelling biodiversity and landuse urban growth agriculture and nature in a wetland areardquo Ecological Economics 51(3-4)201ndash216 doi101016jecolecon200404011 3 1 11

Eskinasi M Rouwette E (2004) ldquoSimulating the urban transformation process in the Haaglan-den region the Netherlandsrdquo in System Dynamics Conference Proceedings (CD-ROM) (Eds)Kennedy M Winch GW Langer RS Rowe JI Yanni JM 22nd International SystemDynamics Conference held in Oxford England July 25 ndash 29 2004 Albany NY (System Dy-namics Society) Related online version (cited on 2 April 2009)httpwwwsystemdynamicsorgconferences2004indexhtm 3 1 19

Ettema D de Jong K Timmermans HJP Bakema A (2007) ldquoPUMA Multi-Agent Mod-elling of Urban Systemsrdquo in Modelling Land-Use Change Progress and Applications (Eds)Koomen E Stillwell J Bakema A Scholten HJ pp 237ndash258 Dordrecht (Springer) doi101007978-1-4020-5648-2 3 1 13

Forrester JW (1969) Urban Dynamics Cambridge MA (MIT Press) 3 1 14 18

Geist HJ Lambin EF (2002) ldquoProximate Causes and Underlying DrivingForces of Tropical Deforestationrdquo BioScience 52(2) 143ndash150 doi1016410006-3568(2002)052[0143PCAUDF]20CO2 5

Geist HJ Lambin EF (2004) ldquoDynamic Causal Patterns of Desertificationrdquo BioScience 54(9) 817ndash829 doi1016410006-3568(2004)054[0817DCPOD]20CO2 5

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

42 Dagmar Haase and Nina Schwarz

Geurs KT van Wee B (2004) ldquoLand-usetransport Interaction Models as Tools for Sustainabil-ity Impact Assessment of Transport Investments Review and Research Perspectivesrdquo EuropeanJournal of Transport and Infrastructure Research 4(3) 333ndash355 URL (cited on 8 April 2009)httpwwwejtirtbmtudelftnlissues2004_03pdf2004_03_05pdf 13

Haase D Haase A (2008) ldquoDo European social science data serve to feed agent-based simulationmodels on residential mobility in shrinking citiesrdquo in Europe and its Regions The Usage ofEuropean Regionalised Social Science Data (Eds) Grozinger G Matiaske W Spieszlig CKpp 227ndash250 Newcastle (Cambridge Scholars Publishing) 5

Haase D Nuissl H (2007) ldquoDoes urban sprawl drive changes in the water balance and policyThe case of Leipzig (Germany) 1870ndash2003rdquo Landscape and Urban Planning 80(1ndash2) 1ndash13doi101016jlandurbplan200603011 5

Haghani A Lee SY Byun JH (2003a) ldquoA System Dynamics Approach to Land Use Trans-portation System Performance Modeling Part I Methodologyrdquo Journal of Advanced Trans-portation 37(1) 1ndash41 3 1 5

Haghani A Lee SY Byun JH (2003b) ldquoA System Dynamics Approach to Land Use Trans-portation System Performance Modeling Part II Applicationrdquo Journal of Advanced Trans-portation 37(1) 43ndash82 3 1 5

Hansen HS (2007) ldquoAn Adaptive Land-use Simulation Model for Integrated Coastal Zone Plan-ningrdquo in The European Information Society Leading the Way with GEO-information (Eds)Fabrikant SI Wachowicz M 10th Conference of the Association of Geographic InformationLaboratories for Europe (AGILE) held in Aalborg Denmark May 8 ndash 11 2007 Lecture Notes inGeoinformation and Cartography pp 35ndash53 doi101007978-3-540-72385-1 Berlin New York(Springer) 4

Heinrichs D Kabisch S (2006) ldquoRisikolebensraum Megacity Strategien fur eine nachhaltigeEntwicklung in Megastadten und Ballungszentrenrdquo GAIA 15(2) 157ndash159 11

Hunt JD Kriger DS Miller EJ (2005) ldquoCurrent Operational Urban Land-usendashTransport Modelling Frameworks A Reviewrdquo Transport Reviews 25(3) 329ndash376 doi1010800144164052000336470 12 13

Iacono M Levinson D El-Geneidy A (2008) ldquoModels of Transportation and Land UseChange A Guide to the Territoryrdquo Journal of Planning Literature 22(4) 323ndash340 doi1011770885412207314010 13

Johnson MP (2001) ldquoEnvironmental impacts of urban sprawl a survey of the literature andproposed research agendardquo Environment and Planning A 33(4) 717ndash735 doi101068a332711

Kasanko M Barredo JI Lavalle C McCormick N Demicheli L Sagris VBrezger A (2006) ldquoAre European cities becoming dispersed A comparative analysisof 15 European urban areasrdquo Landscape and Urban Planning 77(1ndash2) 111ndash130 doi101016jlandurbplan200502003 11

Lambin EF Geist HJ (Eds) (2006) Land-Use and Land-Cover Change Local ProcessesGlobal Impacts Global Change - The IGBP Series Berlin (Springer) 11

Landis JD Zhang M (1998a) ldquoThe second generation of the California urban futures modelPart 1 Model logic and theoryrdquo Environment and Planning B Planning and Design 25(5)657ndash666 doi101068b250657 3 1 7

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 43

Landis JD Zhang M (1998b) ldquoThe second generation of the California urban futures modelPart 2 Specification and calibration results of the land-use change submodelrdquo Environment andPlanning B Planning and Design 25(6) 795ndash824 doi101068b250795 3 1 7

Landis JD Monzon JP Reilly M Cogan C (1998) Development and Pilot Application of theCalifornia Urban and Biodiversity Analysis (CURBA) Model Berkeley (University of Californiaat Berkeley) Related online version (cited on 2 April 2009)httpiurdberkeleyedusitesdefaultfilespubsMG98-01pdf 3 1 8

Liu J Dietz T Carpenter SR Alberti M Folke C Moran E Pell AN Deadman PKratz T Lubchenko J Ostrom E Quyang Z Provencher W Redman CL SchneiderSH Taylor WW (2007) ldquoComplexity of Coupled Human and Natural Systemsrdquo Science317 1513ndash1516 doi101126science1144004 11 2

Loibl W Totzer T Kostl M Steinnocher K (2007) ldquoSimulation of Polycentric Urban GrowthDynamics Through Agentsrdquo in Modelling Land-Use Change Progress and Applications (Eds)Koomen E Stillwell J Bakema A Scholten HJ pp 219ndash235 Dordrecht (Springer) doi101007978-1-4020-5648-2 3 1 16

Lutz W Sanderson W Scherbov S (2001) ldquoThe end of world population growthrdquo Nature412 543ndash545 doi10103835087589 11

Mankiw NG Weil DN (1989) ldquoThe baby boom the baby bust and the housing marketrdquoRegional Science and Urban Economics 19(2) 235ndash258 doi1010160166-0462(89)90005-7 31 4

Millennium Ecosystem Assessment (2005) ldquoEcosystems and Human Well-being Synthesisrdquo MASynthesis Reports Washington DC (Island Press) Related online version (cited on 2 April2009)httpwwwmillenniumassessmentorgensynthesisaspx 11

Miller EJ Hunt JD Abraham JE Salvini PA (2004) ldquoMicrosimulating urban systemsrdquoComputers Environment and Urban Systems 28(1-2) 9ndash44 doi101016S0198-9715(02)00044-3 12 1 10

Moeckel R Schwarze B Wegener M (2006) ldquoDas Projekt ILUMASS ndash Mikrosimulation derraumlichen demografischen und wirtschaftlichen Entwicklungrdquo in Integrierte Mikro-Simulationvon Raum- und Verkehrsentwicklung Theorie Konzepte Modelle Praxis (Ed) BeckmannKJ Tagungsband des 9 Aachener Kolloquium lsquoMobilitat und Stadtrsquo (AMUS 2006) 18 ndash 19September 2008 vol 81 of Schriftenreihe Stadt Region Land pp 53ndash61 Aachen (RWTH) 1 9

Nijkamp P Van Wissen L Rima A (1993) ldquoA Household Life Cycle Model for Residen-tial Relocation Behaviourrdquo Socio-Economic Planning Sciences 27(1) 35ndash53 doi1010160038-0121(93)90027-G 3 1 45 3

Nuissl H Haase D Lanzendorf M Wittmer H (2008) ldquoEnvironmental impact assessment ofurban land use transitions ndash A context-sensitive approachrdquo Land Use Policy 26(2) 414ndash424doi101016jlandusepol200805006 11

Onsted JA (2002) SCOPE A Modification and Application of the Forrester Model to the SouthCoast of Santa Barbara County Masterrsquos Thesis University of California Santa Barbara SantaBarbara 3 1 15

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

44 Dagmar Haase and Nina Schwarz

Parker DC Manson SM Janssen MA Hoffmann MJ Deadman P (2003) ldquoMulti-AgentSystems for the Simulation of Land-Use and Land-Cover Change A Reviewrdquo Annals of theAssociation of American Geographers 93(2) 314ndash337 doi1011111467-83069302004 3

PRB (2007) ldquo2007 World Population Data Sheetrdquo Washington DC (Population ReferenceBureau) Related online version (cited on 2 April 2009)httpwwwprborgPublicationsDatasheets20072007WorldPopulationDataSheetaspx 11

Ramankutty N Graumlich L Achard F Alves D Chhabra A DeFries R Foley JA GeistHJ Houghton R Klein Goldewijk K Lambin E Millington A Rasmussen K Reid RTurner II BL (2006) ldquoGlobal land cover change recent progress remaining challengesrdquo inLand-Use and Land-Cover Change (Eds) Lambin EF Geist HJ pp 9ndash39 Berlin New York(Springer) 11

Raux C (2003) ldquoA systems dynamics model for the urban travel systemrdquo in European TransportConference 2003 Proceedings of ETC 2003 8 ndash 10 October 2003 Strasbourg (CD-ROM) Lon-don (Association for European Transport) Related online version (cited on 2 April 2009)httphalshsarchives-ouvertesfrhalshs-00092186en 3 1 20

Ravetz J (2000) City Region 2020 Integrated Planning for a Sustainable Environment London(Earthscan) 11

Salvini PA Miller EJ (2005) ldquoILUTE An Operational Prototype of a Comprehensive Mi-crosimulation Model of Urban Systemsrdquo Networks and Spatial Economics 5(2) 217ndash234 doi101007s11067-005-2630-5 3 1 10

Sanders P Sanders F (2004) ldquoSpatial urban dynamics A vision on the future of urban dynamicsForrester revisitedrdquo in System Dynamics Conference Proceedings (CD-ROM) (Eds) KennedyM Winch GW Langer RS Rowe JI Yanni JM 22nd International System DynamicsConference held in Oxford England July 25 ndash 29 2004 Albany NY (System Dynamics Society)Related online version (cited on 2 April 2009)httpwwwsystemdynamicsorgconferences2004indexhtm 3 1 14

Schaldach R Priess JA (2008) ldquoIntegrated Models of the Land System A Review of ModellingApproaches on the Regional to Global Scalerdquo Living Rev Landscape Res 2(1) URL (cited on2 April 2009)httpwwwlivingreviewsorglrlr-2008-1 13

Silva EA Clarke KC (2002) ldquoCalibration of the SLEUTH urban growth model for Lis-bon and Porto Portugalrdquo Computers Environment and Urban Systems 26(6) 525ndash552 doi101016S0198-9715(01)00014-X 3 1 17

Sterman JD (2000) Business Dynamics System Thinking and Modeling for a Complex World Boston (IrwinMcGraw-Hill) 3

Strauch D Moeckel R Wegener M Grafe J Muhlhans H Rindsfuser G Beckmann KJ(2003) ldquoLinking Transport and Land Use Planning The Microscopic Dynamic Simulation ModelILUMASSrdquo in Proceedings of the 7th International Conference on GeoComputation Universityof Southampton United Kingdom 8 ndash 10 September 2003 Leeds (University of Leeds) URL(cited on 2 April 2009)httpwwwgeocomputationorg2003 3 1 9

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2

Models on HumanndashNature Interactions 45

Timmermans HJP (2006) ldquoThe saga of integrated land use and transport modeling How manymore dreams before we wake uprdquo in Moving Through Nets The Physical and Social Dimensionsof Travel (Ed) Axhausen KW Selected papers from the 10th International Conference onTravel Behaviour Research Lucerne Switzerland 10 ndash 15 August 2003 pp 219ndash248 Oxford(Elsevier) 12 13

United Nations (2008) ldquoWorld Urbanization Prospects The 2007 Revisionrdquo New York (UnitedNations Department of Economic and Social Affairs Population Division) Related online ver-sion (cited on 2 April 2009)httpwwwunorgesapopulationpublicationswup20072007wuphtm 11

United Nations (2009) ldquoWorld Population Prospects The 2008 Revisionrdquo New York (United Na-tions Department of Economic and Social Affairs Population Division) Related online version(cited on 2 April 2009)httpwwwunorgesapopulationpublicationswpp2008 11

US EPA (2000) ldquoProjecting Land-Use Change A Summary of Models for Assessing the Effectsof Community Growth and Change on Land-Use Patternsrdquo EPA600R-00098 Cincinnati OH(US Environmental Protection Agency Office of Research and Development) Related onlineversion (cited on 2 April 2009)httpfacultywashingtonedupwaddellModelsREPORTfinal2pdf 13

Verburg PH (2006) ldquoSimulating feedbacks in land use and land cover change modelsrdquo LandscapeEcology 21(8) 1171ndash1183 doi101007s10980-006-0029-4 12 14

Verburg PH Overmars KP (2007) ldquoDynamic Simulation of Land-use change Trajectories withthe CLUE-s Modelrdquo in Modelling Land-Use Change Progress and Applications (Eds) KoomenE Stillwell J Bakema A Scholten HJ pp 321ndash335 Dordrecht (Springer) doi101007978-1-4020-5648-2 3 1 6

Verburg PH Schot PP Dijst MJ Veldkamp A (2004) ldquoLand use change modelling currentpractice and research prioritiesrdquo GeoJournal 61(4) 309ndash324 doi101007s10708-004-4946-y13

Waddell P (2006) ldquoUrbanSim ndash Status and Further Developmentrdquo in Integrierte Mikro-Simulation von Raum- und Verkehrsentwicklung Theorie Konzepte Modelle Praxis (Ed)Beckmann KJ Tagungsband des 9 Aachener Kolloquium lsquoMobilitat und Stadtrsquo (AMUS 2006)18 ndash 19 September 2008 vol 81 of Schriftenreihe Stadt Region Land pp 81ndash89 Aachen (RWTH)1 21

Waddell P Ulfarsson G (2004) ldquoIntroduction to urban Simulation Design and Development ofOperational Modelsrdquo in Handbook of Transport Geography and Spatial Systems (Eds) StopherP Button KJ Haynes KE Hensher DA vol 5 of Handbooks in Transport pp 203ndash236Amsterdam Boston (Elsevier) Related online version (cited on 2 April 2009)httpwwwurbansimorgpaperswaddell-ulfarsson-ht-IntroUrbanSimulpdf 12

Waddell P Borning A Noth M Freier N Becke M Ulfarsson G (2003) ldquoMicrosimulation ofUrban Development and Location Choices Design and Implementation of UrbanSimrdquo Networksand Spatial Economics 3(1) 43ndash67 doi101023A1022049000877 Related online version (citedon 2 April 2009)httpwwwurbansimorgpapersUrbanSim_NSE_Paperpdf 3 1 21

Living Reviews in Landscape Researchhttpwwwlivingreviewsorglrlr-2009-2