Spatial Decision Support Systems for policy support in urban planning contexts Roel Vanhout P.O. Box 463 6200 AL Maastricht The Netherlands www.riks.nl
About RIKS w Independent research institute w Founded in the 1980 s to focus on Artificial intelligence E-learning Geographic information systems which eventually led to land use modeling w Focus areas Land use modeling Model integration Integrated modeling for policy support RIKS Research Institute for Knowledge Systems 2
Content w Spatial Decision Support Systems and Integrated Spatial Planning: what and how? w Land use modeling as a central component of the (I)SDSS w Application through traditional scenarios and storyline and simulation approach w Case studies: Province of Utrecht, the Netherlands PRELUDE (EEA) New infrastructure in Puerto Rico 3
Content w Spatial Decision Support Systems and Integrated Spatial Planning: what and how? w Land use modeling as a central component of the (I)SDSS w Application through traditional scenarios and storyline and simulation approach w Case studies: Province of Utrecht, the Netherlands PRELUDE (EEA) New infrastructure in Puerto Rico 4
Integrated Spatial Planning is w An explicitly spatial - temporal domain w Multi-disciplinary, multi-sectoral, and multiple spatial agents w Policy-interventions cause irreversible change w w Policy-interventions require resources this requires a set of models, methods, techniques, data,, available in an integrated environment rather than in isolation 5
What is a Policy Support System? w Computer-based information system w Supports - not replaces - some or all phases of the decision-making process w Interactive and user-friendly w Facilitates analysis, learning and communication w Employs complex and weakly-structured decision contexts (Adapted from M. Scott Morton, 1971 and George M. Marakas, 1999) 6
Weakly-structured decision context Uncertainty weakly structured problem unstructured problem structured problem weakly structured problem Water resources Structured problem: What is the diameter of the aqueducts required if we want to transfer 50.000 hectoliters per day? Conflicts Weakly structured problem: Can we guarantee enough water of high quality at every day of the year for the next 15 years? Unstructured problem: How should we manage the water resources among the different (After actors Hoppe and under Peterse, climatic 1998 in change? Van Delden, 2000) 7
Content w Spatial Decision Support Systems and Integrated Spatial Planning: what and how? w Land use modeling as a central component of the (I)SDSS w Application through traditional scenarios and storyline and simulation approach w Case studies: Province of Utrecht, the Netherlands PRELUDE (EEA) New infrastructure in Puerto Rico 8
Land use modelling with Metronamica Models at 3 coupled spatial scales National, Portugal in EU Regional, 5 NUTS 2 regions Local, 356000 cells 25ha 9
Regional demands Suitability Land use change over time & Land use at & Interaction time T+1 weights & Zoning Transition Rule Cells change to land-use with highest potential until regional demands Time are met. Loop Accessibility Potential for change & = 10
11
The single run is not what counts: Working with uncertainty Probability that the cell is occupied by particular land use as the result of uncertainty in parameter(s). Not 1, but 10, 100,, runs Fluctuating 1, 2,, all parameters Milieu- en Natuureffecten Nota Ruimte RIVM, May 2004 Prepared for VROM (Ministry of public housing and Spatial Planning and the Environment 12
Content w Spatial Decision Support Systems and Integrated Spatial Planning: what and how? w Land use modeling as a central component of the (I)SDSS w Application through traditional scenarios and storyline and simulation approach w Case studies: Province of Utrecht, the Netherlands PRELUDE (EEA) New infrastructure in Puerto Rico 13
Traditional scenario approach Preparation Interviews Construction of preference and zoning maps Workshop 1: Design of alternatives Interactive software tool application Discussion and choice of alternative locations Workshop 2: Calculation of the alternatives Different growth figures Chosen alternatives and LOV-alternative Indicators Analysis of results and discussion 14
Story and Simulation approach Problem defini-on Feedback to stakeholder group Development of dra4 narra-ve scenarios Model runs and analysis of results Model development and/ or applica-on Quan-fica-on of narra-ve scenarios (From van Delden, Riddell, Maier, A methodology for linking exploratory storylines and quantitative modelling in support of long-term integrated planning, Environmental Modeling and Software, upcoming) 15
Content w Spatial Decision Support Systems and Integrated Spatial Planning: what and how? w Land use modeling as a central component of the (I)SDSS w Application through traditional scenarios and storyline and simulation approach w Case studies: Province of Utrecht, the Netherlands PRELUDE (EEA) New infrastructure in Puerto Rico 16
Utrecht - different sectors and their criteria w Environmental hygiene w Social coherence w Economics w Cultural history w Landscape w Traffic and transportation w Ecology & nature w Water resources w Soil 17
Utrecht - result of weighing preference maps of different sectors for industry 18
Utrecht - designing alternatives for housing Alternative 1 Alternative 2 19
Utrecht - calculation of alternatives Alternative 1 Alternative 2 Alternative Environment Explorer 20
Utrecht - indicators Open space Job potential 21
Story and Simulation approach (SAS) for EEA (in PRELUDE) Qualitative scenarios written by stakeholders 1 st meeting Consistent storylines and model calculations Scenarios translated in qualitative driving forces and model parameters 2 nd meeting 3 rd meeting Comparison by stakeholders of the storylines with the model output Qualitative scenario output translated into quantitative description Quantitative modelling of scenario s EU-level Quantitative modelling of scenario s Regional-level 22
SAS in PRELUDE - storylines RIKS Research Institute for Knowledge Systems 23
SAS in PRELUDE - storylines RIKS Research Institute for Knowledge Systems 24
SAS in PRELUDE modeling results RIKS Research Institute for Knowledge Systems 25
SAS in PRELUDE modeling results RIKS Research Institute for Knowledge Systems 26
4 October 2007 27
Puerto Rico: construction PR66 2015 PR66 1990 2015 28
Puerto Rico: probability for urbanization
Others w Other examples: Assessment of desertification risks in the Mediterranean Natural hazard (bush fires, flooding, earthquakes, ) risks in Australia (South Australia) and New Zealand (Christchurch, Auckland, Wellington) Effects of the Common Agricultural Policy in the EU (LUMOCAP) More transport / land use interaction in Saudi Arabia, South Africa, Australia, 30
Thank you & time for questions! (or rvanhout@riks.nl) 31
RIKS Research Institute for Knowledge Systems 32
List of factor maps Land use function Utrecht - the preference maps Factor map Giving a value to the legend items Preference of the sector 33
Utrecht - combining the preference maps Housing Services Industry 34
Utrecht - area to be planned Current locations Fixed plans until 2005 Plans 2005-2015 Expansion 2005-2015 Housing 16825 ha (673 cells) 1350 ha (54 cells) 1425 ha (57 cellen) 1300 ha (52 cellen) Offices 375 ha (15 cells) 175 ha (7 cells) 125 ha (5 cellen) 50 ha (2 cellen) Industry 3625 ha (145 cells) 175 ha (7 cells) 375 ha (15 cellen) 300 ha (12 cellen) 35