Scenarios and Indicators Supporting Urban Regional

Available online at www.sciencedirect.com

Procedia Social and Behavioral Sciences 21 (2011) 243–252

International Conference: Spatial Thinking and Geographical Information Sciences 2011

Scenarios and Indicators Supporting Urban Regional Planning Petrov Laura Oanaa*, Shahumyan Harutyuna, Williams Brendana, Convery Sheilaa a

University College Dublin Urban Institute, Richview, Belfield, Dublin 4, Ireland

Abstract For European urban regional planning, the Impact Assessment tools are used to understand the current situation and to explore concerns about the future of the world. This paper presents the results of scenario modelling and relevant indicators for urban and regional development. Within Europe, Ireland is experiencing one of the severest recessions. It has undergone substantial changes over the past three decades: recession in the 1980s, boom in the 1990s and economic collapse at the end of the 2000s [1]. Over these periods there have been substantial land use changes and population variation [2]. Particularly, the paper investigates the Greater Dublin Region of Ireland where urban development has been intensive and poorly controlled, leading to changes in its spatial configuration and particularly the preponderance of a sprawl type pattern of development. These initial results provide also a platform for a methodology to be used by the two communities of scientists and stakeholders/policy-makers in the application of scenario technique, modeling and indicators approaches towards developing solutions to real world environmental and land use management problems. Proposals for its future development and suggestions for further research are explored.

© 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Yasushi Asami. Keywords: European urbanisation and sustainable development; land use change and impact assessment tools; land use modelling scenarios; urban regional indicators; regional strategic planning, MOLAND

1. Introduction ‘Landscapes are on the political agenda today’ [3]. In 1986, the Single European Act introduced environment considerations for the first time as an integral part of European policy processes. In 1999, the Treaty of Amsterdam reinforced the concerns about environment at European level and therefore, in all sectorial policies environment protection was integrated [4]. Then, in 2002, the Impact Assessment (IA) was introduced by the European Commission (EC) for making policy development more transparent and improving the ‘quality’ of European policies [5]. Assessments of land use changes have initially been accepted only in a qualitative manner as scenarios for future land use [6], but now land use change scenarios together with quantitative analysis and spatial models are tools which are part of European land use and regional planning policies in the IA.

*

Corresponding author. Tel.: +352-661124726; fax: +353-1-716-2690. E-mail address: [email protected].

1877–0428 © 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Yasushi Asami doi:10.1016/j.sbspro.2011.07.012

244

Petrov Laura Oana et al. / Procedia Social and Behavioral Sciences 21 (2011) 243–252

The MOLAND model used for this work is a tool to support decision-making in policy contexts. It was initiated at the Joint Research Centre (JRC) in 1998 as MURBANDY (Monitor Urban Dynamics) project, later followed-up by the MOLAND (Monitoring Land Use/Cover Dynamics) project having the objective to monitor the development of urban areas and identify trends at the European scale. It is directly relevant to several environmental topics at the EU level, supports the initiatives on Environmental Impact Assessment and Strategic Environmental Assessment and addresses specifically the issues mentioned in the European Spatial Development Perspective (ESDP). MOLAND also contributed to the preparation and definition of the Thematic Urban Strategy of the 6th Environmental Action Plan of the EU [7]. Currently at JRC it is the main tool for Land Use Modeling Platform project that supports policy needs of different services of the Commission, for ex-ante assessments and more specific impact assessments (http://moland.jrc.ec.europa.eu). Moreover, the efforts to develop indicators in the management and sustainable development of urban areas, land use and environment have strongly increased since the beginning of the 1990s led by intergovernmental processes ((Organization for Economic, Cooperation and Development (OECD), [8][9]), EEA and EUROSTAT, the statistical office of the European Union [10][11][12]. For example, EEA developed European common indicators to be used at the local level. Then, within MURBANDY/MOLAND projects a number of indicators at the city and regional level were tested [13][14]. The goals of this paper are to analyses alternative future scenarios for regional strategic planning and to examine examples of critical indicators for policy evaluation purposes for Greater Dublin Region (GDR) of Ireland. The work also tests a case study where scenarios and the functionality of a model (MOLAND, in this case study) links the skills and knowledge of scientists and stakeholders/policy-makers in order to develop regionally adapted land use storylines and modify/improve the model simulation. 2. Methodological framework 2.1. Study area As mentioned above the GDR was chosen as the study area for this research. It consists of Dublin (including the city and adjacent areas), Meath, Kildare, Wicklow and Louth counties (Fig. 1). During the last forty years, Ireland experienced unprecedented social, economic and demographic change, particularly in the 1980s, 1990s and 2000s which represent key periods of change in the country and in Dublin city and surroundings [15][16][2]. The economic and population growth in the GDR led to a rapid urban growth influenced by a number of push factors such as high house prices, small size apartments and an inadequate transport system combined with weak planning which rapidly drove urban development outwards into the urban fringes of the Dublin Region [17][2].

Fig.1. The GDR highlighted in the map of Ireland.


2.2. Scenario building Four scenarios of the GDR regional development were produced by early-stage and senior scientists and stakeholders/policy-makers during the summer school and workshop in 2009 at University College Dublin (UCD) Urban Institute Ireland, Dublin. This work was stimulated by previous studies on the GDR [18][2] and/or Greater Dublin Area (GDA) which is one of the case studies of the MURBANDY project of the EC, JRC [19]. The building of scenarios (‘What If?’) in the context of policy option for regional planning and governance turned out to be very appealing and stimulated the discussions more than graphs and text. By using five driving forces (population, economic trends, urbanization, transport and overall trends), the following qualitative scenarios were developed by the participants: “Business as usual” (BU), “Compact development/Environmental friendly” (CD), “Managed dispersed” (MD) and “Recession” (R). We started with narratives and then quantified the scenarios by translating them into MOLAND parameters. In the two meetings a first ‘draft’ of future land use maps of the GDR in 2026 were created. Additionally, in line with the current and future development challenges in Ireland and also based on MOLAND capabilities, critical indicators specifically linked to land use, urban growth and sprawl, and urban/regional development were discussed. Many stakeholders/policy-makers are familiar with scenarios work, but less with spatial modeling. Therefore, the scientists and stakeholders/policy-makers worked together not only in building scenarios and selecting them, but also in the practical application of the MOLAND model. The idea of this exercise was to bridge the two communities in order to collaborate around spatial models and produce future land use maps which should have a clear and accepted interpretation, be robust, statistically validated and respond to policy interventions. 2.3. Scenarios The BU scenario explores the further development of urban patterns emerging before the economic crisis whereas the R scenario focuses on future urban development due to recession, including a recovery by 2016. The CD scenario is important in demonstrating less pressure on natural land uses, exploring urban growth and urban/regional development in the frame of a strong environmental protection policy. In the MD scenario we investigated in more details the growth and sprawl of rural town and villages in open countryside particularly along the Dublin-Belfast motorway. The realization of this scenario is greatly facilitated by the planning regime which imposes few constraints on the conversion of agricultural areas to low-density housing areas [2]. 2.4. The MOLAND model and data available How the MOLAND model works and how the storylines are translated into MOLAND parameters, as well as how it was calibrated for GDR are described in several publications such as White et al., [20]; Barredo et al, [21][22]; Petrov et al, [23]; Shahumyan et al, [24][25]. Therefore, here we focus on presenting data applicable. MOLAND takes as input several geo-referenced datasets for the simulation of urban areas and regions: Land use: The land use maps in MOLAND adopt the CORINE land cover classification, with a fourth more detailed level of nomenclature added for artificial surfaces [26]. The GDR land use maps of 2000 and 2006 were produced by ERA-Maptec Ltd as part of the Urban Environment Project (UEP, 20062010, www.uep.ie). Figure 2 represents the land use map of 2006 which was the starting year of our scenario simulation. The county boundary and electoral division maps were obtained from Ordnance Survey Ireland (OSI) national mapping agency (Permit No. MP009006). Transport: Existing road and rail datasets from the Dublin Transportation Office and the National Roads Authority were used in all scenarios. However, each one of the scenarios required different link updates derived from the planned network (www.transport21.ie). Suitability: To implement the effects described in each scenario, some modifications were made to the suitability maps for residential and industrial classes. The suitability of key towns was kept relatively high while in the rural areas it was decreased. Thus, the main differences of the suitability maps are: the highly suitable towns in Dublin-Belfast transport corridor for the MD scenario; and a restricted

245

246


zone (2km buffer along coastline) for future development for CD scenario. For the BU and R scenarios a default suitability map was used.

Fig. 2. Land use map of GDR in 2006.

Zoning: For the GDR application initially attempts were made to apply direct translation of the statutory development plans into zoning maps for MOLAND. This process was however tedious and time consuming due to poor data quality, incomplete and inconsistent data and lack of homogeneity within counties. As a result, a one-to-one mapping of the zoning maps into the MOLAND could not be incorporated. An alternative approach was implemented, whereby zoning maps were developed


including special protection and conservation areas as well as national heritage areas. The same zoning map was applied for all residential and other urban (activity) land use classes. Socio-Economic Data: At the regional level, MOLAND requires socio-economic data for each of the modeled counties. For the simulation starting year 2006 we have used census population and employment data provided by the Irish Central Statistics Office (CSO). For population 2026, the “M2F1 Traditional” projection was used for BU, CD and MD scenarios, assuming international migration declining with constant fertility and a return to the traditional pattern of internal migration by 2016; “M0F1 Traditional” projection was used for R scenario and considers zero net migration and high fertility [27]. Job numbers for BU scenario were extrapolated using population and job figures from 2006 and CSO population projection of 2026, assuming that distribution of jobs in industry, commerce and services will remain the same as it is in 2006, and that the absolute numbers will increase in accordance to population growth by 2026. The MD scenario suggests a steady increase of economic growth by 2026. For CD scenario, 30% more jobs are projected compared with MD scenario. The employment projections for R scenario are based on the Economic and Social Research Institute (ESRI, www.esri.ie) where relatively a short recession is followed by a strong recovery [28]. 3. Results 3.1. Land use change scenario comparison Development patterns differ between scenarios not only by urbanization strength but also due to different spatial policies (e.g. preserving the land along the coast, policies aiming at compact urbanization) and suitability. Figure 3 compares the existing urban development in 2006 and 2026 (green), existing in 2006, but not in 2026 (blue), and urban areas developed from 2006 to 2026 (red) for the four scenarios. Here, it can clearly be observed that County Dublin has significant urban areas by 2006. Data analyses show that the urban area of County Dublin has been dominant by 58% in 2006, followed by Kildare (14%), Meath (10%), Wicklow (9%) and Louth (8%). It can also be seen that during 2006-2026 period, the highest increase of urban areas is in BU (268%) and MD (269%) scenarios in County Meath, followed by Wicklow, Kildare and Louth. In 2006, the GDR with total area of 780908 ha contains 4.6% residential, 0.7% industrial, 0.2% commercial and 0.3% service areas. The differences between scenarios are mainly due to population and employment projections, and also environmental policies adopted. As a result, in 2026 the GDR is foreseen to have 9.2% residential areas, 1% industrial, 0.4% commercial and 0.4% service areas in BU scenario. Then, in MD scenario residential and commercial areas are the same as for the BU scenario while industrial and service areas slightly increase to 1.1% and 0.5% respectively. The environmental friendly CD scenario comprises 8.5% residential, 1.4% industrial, 0.6% commercial and 0.6% service areas. Noteworthy, is the R scenario with 7.6% residential areas, 0.8% industrials, 0.2% commercials and 0.3% services. 3.2. Indicators In the context of environmental indicators developed by EEA and also based on urban sprawl which became a major topic of environmental concern [13], MOLAND outputs can be used to calculate indicators for urban and regional development that allows comparisons in Europe and internationally [29][7]. Within GDR, the greatest transformations occurred in and around Dublin which was the principal beneficiary of the ‘Celtic Tiger’ growth [30][31], therefore, we further take a closer look to County Dublin. 3.2.1. Land use indicator The ‘land use transformation versus time’ is an indicator of land use change over a period of time, involving comparisons of each cover type for each cell between the start and end of the model run. Between 2006 and 2026, the residential areas of County Dublin expand by almost 50% in all scenarios,

247

248


except in R scenario where it increases only by 32%. Residential continuous areas in County Dublin varies from 45% in CD scenario to 21% in R scenario; meanwhile residential sparse areas rise by 150% in BU and MD scenarios, 118% in R and 109% in CD scenario. Industrial areas increase by 52% in CD scenario and diminish their activity area (abandoned areas) by -12% in R scenario. Similar trends are seen for commercial and service areas, increasing significantly in CD scenario (over 60%) and showing the lowest increase in R scenario.

Fig. 3. The alternative urban development of GDR for 2026-year.


3.2.2. Urban growth indicator The ‘urban growth’ indicator as implied by its name, is an inherently dynamic spatial phenomenon. By 2026, in County Dublin the residential areas represent between 74 and 91% of the urban area. The maximum growth of residential areas is in CD scenario. Figure 4 shows that residential sparse growth is more than 80% in all scenarios and over 90% in CD and R scenarios. Residential continuous growth is about 63% in R scenario and 65% in the other three scenarios.

Fig. 4. Urban growth in the four scenarios of 2026-year for County Dublin

Calculations show that the loss of heterogeneous agricultural and pasture areas is the highest in County Dublin. Table 1 below shows the total natural areas loss within each scenario. Table 1. Urban areas in 2026 and loss of natural areas due to urban growth from 2006 to 2026. Change by Scenarios (%)

Total Urban areas Loss of total Natural areas Dublin Co. Urban areas Loss of Dublin Co. Natural areas

BU

CD

MD

R

86.2 -5.6 41.8 -21.1

89.2 -5.8 53.4 -26.9

89.3 -5.8 45.6 -23

51.8 -3.4 20.8 -10.5

3.2.3. Urban sprawl indicator The ‘urban versus active land uses’ within each scenario analyses the land use structure of the region which gives an impression of where the changes are taking place [32]. For County Dublin, the scenarios of 2026 highlights that the residential areas are dominant compared to the other classes. For example, detailed calculations indicate that residential sparse increase from 8% in 2006 to 30% (in BU scenario), 28% (in MD scenario) and 17% (in CD scenario) in 2026.

249

250


4. Discussions and conclusion The MOLAND application ensures that all relevant aspects of a scenario (demographic, economic, urban growth and sprawl, suitability and zoning policy) are integrated into a consistent framework. The approach of using MOLAND in this work is not only to analyses land use dynamics and design scenario development, but also to explore further the capabilities of this tool. In this larger context, using the outputs of MOLAND we additionally compiled indicators to provide stakeholders/policy-makers with information in the form of indicators to allow them to view and compare the results with different regions. The results of ‘land use’ indicator show that in the CD scenario of County Dublin, industrial, commercial and service activities are the highest. But what it is notably in CD scenario are the residential sparse areas which rise three times more than continuous. Figures 3 shows that total urban area had expanded largely in Dublin city and county before 2006 and the ‘urban growth’ indicator suggests still a dramatic increase of urban area in County Dublin in the three (BU, CD and MD) scenarios, reaching maximum in CD scenario (Fig.4). Nevertheless, the results of the CD scenario suggest that restricting the future development within 2 km buffer from coastline will help to protect sensitive zones even in face of increasing development. Urban sprawl is expressed mainly through strong growth of sparse/discontinuous residential areas [18]. Kasanko et al [14] show that the city of Dublin has the highest rate of discontinuous residential areas among the fifteen European cities studied. Further, this work shows that the greatest urban sprawl occurs in BU scenario, followed closely by MD scenario, being almost four times more than in 2006. Concerning the loss of natural areas, between 2006 and 2026, the heterogeneous agricultural (in BU, CD and MD scenarios) and pasture (in R scenario) areas are mostly transformed into urban areas in County Dublin. And, as table 1 indicates, the total loss of natural areas are minimum in R scenario. To conclude, the overall urban areas have grown considerably in all four active land uses. By 2026 urban development pressures, particularly caused by residential development will continue be experienced in County Dublin. Concerning the growth of economic activities, they follow the trends expected, achieving the maximum activity in CD scenario and the lowest in R scenario. Regarding the other counties of GDR: County Meath should be guarded by stakeholders/policy-makers regarding residential sparse while County Wicklow for the residential continuous. Industrial area reaches a maximum in County Kildare, commercials in County Louth and services in County Meath. More indicators could be employed, but we opted in this research to employ the key indicators that captured key aspects of GDR ‘condition’, particularly County Dublin. The indicators chosen are both critical in their impact and are easily communicated to stakeholders/policy-makers and public. Finally, we analyzed the alternative future scenarios together with the several indicators to show the kind of information thtat can be produced in order to evaluate more efficiently the effects of alternative urbanization patterns of a region. As shown by the results, the focus is not to analyses in depth every nuance of each scenario and indicator, but to give an insight into how such an approach and collaboration can support stakeholders/policy-makers to reduce the impact of new developments and implicitly achieve a better environment. Such strategy can link scientists and stakeholders/policy-makers to work together not only in the phase of building qualitative scenarios, but also in practical application of spatial modeling. They can share experiences and learn from one another which may lead to more effective mitigation options for a sustainable urban/regional land use planning. The approach can be further improved and be beneficial to many regions in Europe and worldwide.


Acknowledge The authors would like to acknowledge the support of the Environmental Protection Agency under contract 2005-CD-U1-M1. All work undertaken with the MOLAND model for the GDR is under licence from RIKS b.v. and Joint Research Centre under licence no. JRC.BWL.30715.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]

[25] [26]

[27]

Brendan Bartley and Rob Kitchin. Understanding Contemporary Ireland, London: Pluto; 2007. ISBN: 9780745325941 Williams B, Hughes B, Redmond D. Managing an unstable housing market. UCD Urban Institute Ireland Working Paper 2010; Series 10/02 Antrop M. Landscape change and the urbanization process in Europe. Landscape and Urban Planning 2004; 67: 9 – 26. Bainbridge T. The Penguin Companion to European Union. London: Penguin Books Ltd; 2005. CEC. Impact Assessment Guideline. 2005; SEC 791 Petit S, Firbank L, Wyatt B, Howard D. MIRABEL: models for integrated review and assessment of biodiversity in European landscapes” Ambio 2001; 30: 81 – 88. Lavalle, C, McCormick N, Kasanko M, Demicheli L, Barredo J. Monitoring and forecasting the dynamics of European urban areas: The territorial approach as key for urban development strategies. Urbanistica 2003; vol.121, p. 105 – 111 OECD. Better understanding our cities: The role of urban indicators. OECD Publication, Paris; 1997a. OECD. Towards sustainable development: environmental indicators. OECD Publication, Paris; 1997b. EC. Communication from Mr. Almunia to the Members of the Commission on Sustainable Development Indicators to monitor the implementation of the EU Sustainable Development Strategy. 2005a ; SEC 161 final EC. Sustainable Development Indicators for the European Union, Eurostat, Luxembourg; 2005b. EEA. EEA core set of indicators, Guide, EEA Technical report No 1, Copenhagen; 2005. EEA. Towards an urban atlas. Environmental issue report No 30, Office for Official Publications of the European Commission, Luxembourg: 2002, ISBN 92-9167-470-2. http://www.pedz.uni-mannheim.de/daten/edz-bn/eua/02/front.pdf Kasanko M, Barredo JI, Lavalle C, McCormick N, Demicheli L, Sagris V, Brezger A. Are European cities becoming dispersed? A comparative analysis of 15 European urban areas. Landscape and Urban Planning 2006; 77: 111-130. Hourihan K. Urban change in the Republic of Ireland. Cities 1989; 6 (3): 209-226. Kitchen P. Identifying dimensions of urban social change in Dublin – 1986 to 1996. Irish Geography 2002; 35: 156-174. Bannon MJ. The Greater Dublin Region: planning for its transformation and development. In: Killen J, MacLaran A, editors. Dublin: Contemporary Trends and Issues for the Twenty-first Century, Dublin: Geographical Society of Ireland; 1999 p.1-19 EEA. Urban sprawl in Europe, The ignored challenge, EEA report No 10, Copenhagen; 2006. ISSN 1725-9177 White R, Engelen G, Uljee I, Lavalle C, Erlich D. Developing an Urban Land Use Simulator for European Cities. In: Fullerton K, editor. Proceedings of the 5th EC-GIS Workshop, Stresa, Italy; 2000, p. 179 – 190 White R, Engelen G, Uljee I. The use of constrained Cellular Automata for high-resolution modelling of urban land-use dynamics. Environment and Planning B: Planning and Design 1997; 24 (3): 323 – 343. Barredo JI, Kasanko M, McCormick N, Lavalle C. Modelling dynamic spatial processes: simulation of future scenarios through cellular automata. Landscape and Urban Planning 2003; 64 (3): 145 – 160. Barredo JI, Demicheli L, Lavalle C, Kasanko M, McCormick N. Modelling future urban scenarios in developing countries: an application case study in Lagos, Nigeria. Environment and Planning B: Planning and Design 2004; 32: 64 – 84. Petrov LO, Lavalle C, Kasanko M. Urban land use scenarios for a tourist region in Europe: Applying the MOLAND model to Algarve, Portugal. Landscape and Urban Planning 2009; 92: 10-23. Shahumyan, H., White, R., Twumasi, B., Convery, S., Williams, B., Critchley, M., Carty, J., Walsh C., Brennan M. The MOLAND Model Calibration and Validation for Greater Dublin Region. UCD Urban Institute Ireland Working Paper 2009a; Series 09/03 Shahumyan H, Twumasi B, Convery S, Foley R, Vaughan E, Carty J, Walsh C., Brennan M. Data Preparation for the MOLAND Model Application for Greater Dublin Region. UCD Urban Institute Ireland Working Paper 2009b; Series 09/04 McCormick, N., Lavalle, C., Barredo, J. An example of the application of spatial dynamics modeling for evaluating the future impact of land use development decisions. In: Proceedings of the 4th Seminar of the PlanNet Europe Network on Strategic Environmental Assessment (SEA) of Urban Plans and Programs. Ministry of the Brussels-Capital Region, Brussels; 2005, p. 47–55, http://planet.difu.de/2005/proceedings/index.phtml. Central Statistics Office (CSO). Regional population projections 2011-2026, Dublin; 2008.

251

252


[28] Bergin A, Conefrey T, Fitzgerald J, Kearney I. Recovery Scenarios for Ireland. Dublin: The Economic and Social Research Institute Research Series. Research Series No 7; 2009. [29] Engelen G, Lavalle C, Barredo JI, van der Meulen M, White R. The MOLAND modelling framework for urban and regional land-use dynamics, In: Koomen, E, Stillwell, J, Bakema, A, Scholten, HJ, editors. Modelling Land-Use Change, Progress and Applications, Springer: Dordrecht; 2007, p. 297 – 319 [30] Couch C, Leontidou L, Petschel-Held G. Urban Sprawl in Europe. Blackwell: Oxford; 2007. [31] Gkartzios M, Scott M. Countryside, Here I Come: Urban Rural Migration in the Dublin City-Region. Planning and Environmental Policy Research Series 2005; PEP 05/01 [32] Farrow A, Winograd M. Land Use Modelling at the Regional Scale: an input to Rural Sustainability Indicators for Central America. In: Veldkamp A, Lambin E, editors. Modelling land use/cover change from local to regional scales, Agriculture Ecosystems and Environment 2001; 85 (1-3) 249 – 268.