Spatial dynamic modelling for urban scenario planning: A case study of Nanjing,China

Abstract

Although urban scenario planning is widely applied for exploring various directions of urban development, it often has high requirements on the medium of quantitative information analysis and transformation. Thus, this study establishes a method of combining scenario planning with a spatial dynamic planning support system to predict urban growth. Specifically, a scenario-based spatial dynamic modelling method is integrated with the information module of planning policy for better decision support. The integrated modelling method is applied for an actual urban land use planning case of Nanjing, an evolving city in China. The spatial forms of future urban land use are simulated under four different pre-set policy scenarios. The differences in simulated results under multi-criteria restrictions reveal the effectiveness and practical value of the integration approach. The findings of this study provide policymakers with a process-based approach to test and evaluate ‘what-if’ consequences and help stakeholders reach consensus.

Keywords

Scenario planning urban land use planning support system urban spatial dynamics what-if

Introduction

The rapid boost of urbanisation in developing countries, such as China, leads to large urban conversions of natural and agricultural land to non-agricultural use (Foley et al., 2005; Liu et al., 2014; Zhou et al., 2017). This massive scale of land conversion, which is mainly dominated by socio-economic factors (e.g. economic development and urban population growth), has induced serious environmental consequences, thereby threatening urban and regional sustainable development (Foley et al., 2005). On this basis, land use planning as a comprehensive instrument for managing construction land expansion has been highly valued by central and local governments in the overall planning of sustainable, healthy and orderly urban development (Kaiser et al., 1995; Liu et al., 2014). Land use planning is formulated to coordinate the scale and spatial distribution of construction land growth. However, as decision-makers are often forced to make strategic decisions under future uncertainties, their decisions may have negative and irreversible effects on the city (Volkery and Ribeiro, 2009).

The emergence of the scenario planning concept provides a way of acknowledging the uncertainty of future development and various possible development trends (Perveen et al., 2017). Typically, a well-designed urban scenario planning can help urban decision-makers reflect on their assumptions about the impact and adaptability of future urban development strategies and policies (Long et al., 2012; Zegras and Rayle, 2012). Although scenario planning has been widely used in the urban planning field, its practical application still faces challenges. The fundamental reason lies in its complexity and difficulty in integrating quantitative analysis methods and tools and the absence of general guidelines to implement and construct scenarios in urban planning practice (Amer et al., 2013). Therefore, the current scenario planning has a universal theoretical basis while the limitation lies in the lack of translation media for interpretation and implications. This limitation further makes it difficult to reach a consensus among different stakeholders on the assessment of future urban development scenarios and thus lowers its application value. Particularly, as in China, quite often the central and local governments may have uncoordinated future urban development scenarios (Wu, 2015). The latter should be aligned with the former’s development scenarios and should set their own policies to maximise individual economic development under the influence of decentralisation and market reform (Wu, 2015). On this basis, a scientific prediction of future construction land expansion that can accommodate future population and economic growth will be helpful in evaluating the implementation effect of different policy scenarios (Zhao et al., 2014; Zhou et al., 2017), which, in turn, is of great significance for policy improvement.

Planning support systems (PSS) are defined as a decision support tools using geo-information science and technology, which can be developed to provide predictive information on future urban development (Geertman and Stillwell, 2009; Klosterman and Pettit, 2005). A class of PSS with dynamic spatial modelling as its main functionality plays a role in the policy choice and theoretical analysis in scenario planning (Long and Zhang, 2015). Relevant studies point out that using model-based PSS tools can systematise the known approach for scenario construction (Deal and Schunk, 2004; Liu et al., 2017). Specifically, some model-based PSS simplify complex urban development issues, provide a balance between urban land supply and demand for different land uses in different locations while still taking specific policy requirements into account (Klosterman, 1999; Liu and Silva, 2017). However, the limitation is that existing model-based PSS often ignore the possible changes and influences brought about by spatially explicit socioeconomic and environmental factors that largely influence land-use changes (Dendoncker et al., 2007; Liu et al., 2014; Lo and Yang, 2002; Mitsuda and Ito, 2011). Thus, logical models must be formulated and designed according to the characteristics and expected development scenarios of specific city cases to correctly simulate the complexity of the interaction system of urban development drivers and so that spatial dynamic modelling PSS can truly achieve the purpose of planning assistance.

In summary, on the one hand, urban scenario planning needs more accurate quantitative information processing tools in determining the possibility of future development. On the other hand, the spatial dynamic model as a PSS has spatial data processing capacity that needs to combine specific planning scenario objectives and methodologies to enlarge its application value. Therefore, a combination of the modelling PSS method with the theoretical guidance of the scenario planning approach may effectively improve the prediction and assessment of future urban development under different pre-set policy scenarios. Recently, increasing studies have begun to use the spatial dynamic modelling method for urban development scenarios under external factor changes (e.g. Pan et al., 2018; Zheng et al., 2012). However, relatively few studies have focused on urban policy scenarios, especially in terms of their systematization and controllability (Deal and Pan, 2017). To fill this gap, we have established a scenario-based spatial dynamic modelling method with the information transformation module of planning policy for better decision support. Furthermore, this modelling method was adopted specifically for urban land use, which is one of the most incipient and important aspects of comprehensive urban development. On the one hand, urban land use issues can be better implemented by a spatial dynamic modelling PSS. On the other hand, urban land use, as previously highlighted, continues to experience abrupt changes, and its regulation plays a significant role in the overall planning of urban development (Liu et al., 2014; Zhou et al., 2017). Secondly, we carried out a specific city case study to ensure the consistency of the study area’s policy background. In this way, the impact of differences on regional policies is reduced, which leads to further improvement in the accuracy of future prediction results. Using a case study of actual urban land use planning in Nanjing, the following three research questions are explored in this study: (1) How is a spatial dynamic modelling PSS integrated with the scenario planning approach in practical urban planning cases? (2) Can the integrated scenario-based modelling method provide predictive and evaluative information based on different future development scenarios? (3) Can this modelling method help deal with the trade-offs associated with complex development issues and the optimisation of specific planning policies? We aim to contribute to the methodological and theoretical field in establishing a scenario-based dynamic land-use modelling method that is applicable for addressing the ‘what if’ question in urban land use planning. Moreover, we aim to contribute to the practical field in extending the policy analysis module in the method and enriching its policy information responsiveness to help policymakers evaluate and optimise urban development policy strategies.

Related works

Scenario planning approach

Various scenario planning approaches have been developed and implemented in practice in recent decades. These approaches bring about the difference between the general method and the specific operational steps. The different scenario planning methods are highly flexible because of the different ways of processing, and most of them were designed for a particular case and situation (Stojanovic et al., 2014). Schwartz (2012) and Shoemaker (1995) established a frequently and commonly used method. The principle of this method is problem-centric extensions. This method emphasises the definition of the problem and related influencing factors, including stakeholders, change in trends, limitations and some other significant issues. According to the classification of the uncertainty and importance of these factors, Inayatullah (2008) divided them into three categories according to the treatment stages: future thinking, future problems and reconciliation. From these findings, the Institute for Alternative Futures developed an optimised method that combines the cyclic learning of uncertainty with the expectation to determine the preferred future development mode (Bezold, 2009).

Exploring the evolution of the desired goal should be a process that includes the realisation of all reasonable scenarios (Wilson, 1994). According to the various expressions of scenario planning approaches, a goal-driven approach can better help decision-makers create a desirable future rather than deal with future situation issues (Tevis, 2010). This goal-oriented scenario planning is also suitable for embedding PSS. Once the scenario planning approach is defined and standardised, the data and modelling-based PSS can be used within the process to quantify the information of development-driven elements (Xu et al., 2020). Researchers interpret the interconnectedness, unintended consequences and reduce the likelihood of extreme situations by using follow-up methods, such as impact analysis, to deal with the results of scenario planning (Meissner and Wulf, 2013; Raford, 2015). If people only focus on illustrating trade-offs related to complex issues, then the guidelines could also help generate additional realistic public feedback (Berg et al., 2016). In general, the use of scenario planning guidelines can realise not only the prediction of the ‘normative’ scene but also the construction of an ‘exploratory scene’.

Scenario-based spatial dynamic modelling PSS

The application value of the spatial dynamic modelling PSS based on scenario analysis is reflected in the evaluation of the change expectation and possibility (Geertman, 2002; Johnson and Sieber, 2009). With the developments of geographic information system (GIS) and computer-based analytical models, a series of assistive planning tools were developed, which are highly nested in urban scenario planning (Burian et al., 2015; Xiang and Clarke, 2003). They all provide a platform for creating and testing urban development scenarios, for example, the model for the conversion of land use and its effects (CLUE) (Veldkamp and Fresco, 1996), the What-If model (Klosterman, 1999), the Land Use Evolution and Impact Assessment Model (LEAM) (Deal and Schunk, 2004) and the Land-Use Conflict Identification Strategy (LUCIS) (Carr and Zwick, 2007). These models are just many of the few examples of scenario integration. Although various spatial dynamic models differ in their operation and practical application, they are undeniably capable of handling advanced land evolution simulation and modelling on a city and a regional scale.

In recent years, developers, users and researchers of the spatial dynamic modelling PSS have conducted scenario-based case studies on these modelling methods, specifically in urban land use planning and development scenario forecasting. The What-If model is an early example of a PSS that integrates scenario planning concepts (Xiang and Clarke, 2003). Du and Li (2005) discussed, in depth, the application of the What-If model in urban spatial development. According to their description, the model provides a comprehensive operating environment for land demand prediction and future land use allocation. Murtha et al. (2019) applied a comparative study on previous LUCIS modelling results (Carr and Zwick, 2005) and spatial and temporal patterns of urban land use change in Orlando, USA. The study provides an in-depth discussion of geo-design frameworks and future scenario formulation. Zheng et al. (2012) developed a system dynamics model (SDM) that couples with CLUE-S (small regional extent version of CLUE) for land use change simulation in a city-scaled study. Their study also systematically addressed the impact of development scenarios on land expansion demand. This space-based supply and demand relationship has also been proven to exist not only regionally (Yu et al., 2011) but also globally (Chen et al., 2020). Deal and Pan (2017) implemented LEAM to address environmental impacts in policy scenarios. LEAM has been proven as an effective method to identify the ‘what-if’ and ‘so-what’ questions (Pan et al., 2019a). Given LEAM’s active feature of multi-model combination, continuous studies have been performed that help investigators conduct joint project studies, such as the ecosystem service study (Pan et al., 2019b), a hydrology study (Pan et al., 2018) and regional economic prediction (Pan and Deal, 2019). These studies showed that a spatial dynamic modelling PSS has sufficient application in system integration, simulation accuracy and planning application, and its function development trend of scenario analysis has gradually evolved into the basis of the urban case study.

Methodology and theoretical framework

Formulation of research design

If the traditional urban planning is ‘planning for people’, then the PSS combined scenario planning is inclined towards ‘planning with people’ (Geertman and Stillwell, 2012). Relevant stakeholders including policymakers, planners and the public need to be involved in the process of urban planning to realise this bottom-up concept and to provide decision support for the planning implementation (Koontz, 2005). First, an understanding of what the desired future should be (i.e. development vision and expected goals) among these stakeholders is needed to achieve this goal. Second, with the help of the model-based scenario planning approach, we interpret the future development trend by simulating the possible policy scenarios, which are quantified and visualised. Third, whether the development situation meets the desired goals can then be determined by comparing the simulation results under different policy scenarios. Finally, additional information can be obtained through the evaluation of the results, which helps decision-makers optimise the development goals or planning and policy strategies.

Model-based scenario planning approach

Inspired by the goal-oriented scenario planning method developed by Schwartz (2012) and simplification ideas of Kosow and Gaßner (2008), we designed the following conceptual framework of scenario planning for this study – from task analysis to what-if test. The conceptual framework is further specified into four theoretical phases as shown in Figure 1. The blue and green lines indicate the theoretical pathway and technical pathway, respectively. At the object categorization phase, we determine the overall goals and strategies of local urban development. Then, these targets are translated into factors of related drivers, projection and constraints in factor identification, whereas the database management system retrieves the factor-related spatial-temporal data for further processing. We can build land use planning scenarios based on these factors by implementing the SDM. Finally, we can obtain the influence of the consequence through what-if tests, which helps to confirm whether the overall development goal is met and whether the strategies need to be adjusted and whether a scenario needs to be reconstructed.

Figure 1.

Conceptual framework of the scenario planning process.

Scenario building method

In the urban planning context, policy scenarios are never set into an agreement and are revised constantly on the way to the desired future (Barredo et al., 2004). Setting scenarios based on development goals and policy constraints provides policymakers with a rich perspective on future uncertainty. The key to realising this is to involve maximum standpoints from local stakeholders (Chakraborty et al., 2011). However, in scientific research, attempts to collect information through traditional methods (e.g. questionnaire survey and workshops) are faced with the issues of inefficiency and subjectivity. In this study, we chose to directly extract valid information from various types of government planning documents related to Nanjing (e.g. Strategic Planning, Master Planning and Five-Year Plan) from the national, regional and local levels. The formulation of different types of planning at different levels in China involves diverse stakeholders with specific interests. Thus, these different governmental planning documents deliver specific development expectations of the majority of stakeholders (cf. Wu, 2015). Information extraction from planning documents is a method of synthesis that crystalises the development history, expert wisdom and the experience of citizens and representatives. Consequently, we quantified and transformed this information into different scenarios with a box of different development targets – that is, the development restriction area, the total area of urban growth, the growth of the urban population and the employed population.

Case study

Study area

The city of Nanjing is selected as the study area because of its rapid urban growth, the dominance of policy scenarios in urban development and the availability of geo-information and policy documentation. As shown in the city map of Figure 2, the study area Nanjing is located in the southwest corner of Jiangsu province and the heart of the Yangtze River Delta. As the capital of Jiangsu Province in China, Nanjing has witnessed rapid economic growth and rapid expansion of construction land since China’s reform and opening-up which is driven by industrialisation, urbanisation and globalisation (Loo and Wang, 2018). This event becomes an important gate-way for the development of the central and western regions of China. The study area currently has five city districts and six suburban districts, covering a total area of 6597 km². From 1980 to 2015, the construction land in Nanjing increased from 883.97 km² to 1646.78 km², with an average annual growth rate of 2.86%. In addition, the local population has increased from 1.73 million to 8.24 million, with an average yearly growth rate of 10.75%. All these aspects show that Nanjing has made a qualitative leap at the urban scale and in terms of development.

Figure 2.

Land use map of Nanjing, China, for 2015.

Data acquisition

The research data involve Nanjing and China provincial boundary data, Nanjing’s land use data and land use change driving factor data. Most of the data were obtained from the Yangtze River Delta Science Data Center, the National Earth System Science Data Infrastructure and the National Science and Technology Infrastructure of China (NSTI-GEODATA.CN), including Nanjing’s land use classification data with a 30-m resolution. The data of regional land use change drivers include the physical geographic and socio-economic factors, such as a digital elevation model, the road network, employment centre distribution, population centre distribution and restricted construction area data. All data were acquired based on the principle of the most recent release first access. Thus, the 2015 data were used as the input year for the modelling. Table S1 in the Supplemental material shows the detailed description of the year, format, coordinate system and source for the above data.

Scenario-based spatial dynamic modelling

The scenario-based spatial dynamic modelling is composed of two major parts: dynamic land use model and policy scenario analysis module. In the first part, we applied LEAM to establish a specific city’s dynamic land use model. The LEAM is a useful tool for encoding spatial structures and helps generate quantitative analysis information to predict future urban land expansion (Deal et al., 2017). The Supplemental material contains the detailed LEAM model description. Figure 3 depicts the flow of the detailed modelling process. First, the physical geographic and socio-economic data are pre-processed as input data to the dynamic land use model to generate local attractors. Second, the growth driver calibration is then processed to ensure the effectiveness of each attractor. Third, the policy scenario analysis module helps define the no-growth zones, together with calibrated attraction factors and demographic projections, which are used to generate the change probability map. Fourth, various development scenarios are proposed based on the development goals and expectations in the selected government planning documents, which drive the cell’s possibility value thresholding for final change simulation. The dynamic land use modelling is capable of fine-tuning any of the independent drivers and expected growth projections to suit different planning policy scenarios because of its multi-modular design. Thus, this open model framework allows the principles of the scenario planning approach to be applied.

Figure 3.

Flowchart of the scenario-based spatial dynamic modelling.

The policy scenario analysis module takes government planning documents as the core and extracts effective planning policies and development goals for analysis. The specific analysis approach is to transform the quantified development indicators or development goals into factor-related data, including the projection and restriction data. Among them, the projection data include population growth, employment growth and planned road construction, which will ensure the authenticity of the driving factors in modelling the urban land use change at a predicted point in time. Restriction data include water areas, environmental and ecological protection, agricultural land protection and regional restricted construction areas. These data were used as the initial input data of the dynamic land use model, which determines the change effect area during the modelling operation. The dynamic land use model calculates the area of land use change and the change probability of each cell within the area. Thus, we can obtain a certain number of changed areas by probability thresholding. According to the development needs in the government planning documents, the area of the expected transformation of land use is calculated and then superimposed onto various indexes, which provides a basis for changed area selection. Finally, we use the condition judgement method to determine whether the captured changed area meets the government’s development expectations to further ensure the accuracy and effectiveness of the simulated changes.

Policy scenario setting

Policy scenarios are built to be dynamically reactive to ecosystem service impact. These scenarios aim to reflect the concept of socio-natural solutions to urban sustainability and its use in policy formation and planning decisions. Dynamic and replicable policy scenario testing allows for advanced decision support in local and regional planning. Thus, the study on city planning policies and development strategies includes national, regional and local dimensions. According to the National Spatial Planning Master Strategy, the Outline of the Integrated Regional Development of Yangtze River Delta and the Nanjing’s Master Plan for 2035, we summarised the general development goals by priority into the following points: (1) maintain the farmland and ecological preservation areas, (2) define the boundaries of urban development, (3) develop along both sides of the Yangtze River and (4) accelerate urban and rural integration.

We then identified the existing countermeasures that could be used to implement predetermined policy parameters for the planning scenario (as shown in Table 1). We developed the following four scenarios to highlight the trade-offs between economic and ecological aspects while interpreting the urban development goals and policy restrictions of Nanjing: S1 – natural growth mode, S2 – rapid economic growth mode, S3 – ecological and sustainable development mode and S4 – comprehensive and coordinated development mode. All these planning scenarios are interpreted in terms of driver and projection input of the dynamic land use model so that they can correctly be transformed into the land use model scenarios. Table 1 shows the development goals, input constraints and projected population data for each established scenario. For the natural growth mode, we adjusted the projection input to the baseline and removed the limits on the restricted areas except for the water areas. For the rapid economic growth mode, the population and employment centres were re-projected based on a faster growth scenario, and upcoming major roads were added to the road network. For the ecological and sustainable development mode, the ecological reserves and urban concentrated agricultural land were incorporated into the urban no-growth zone. The comprehensive and coordinated development mode combines the first three scenarios but with a more balanced model projection. All data were manually digitised into geo-information data according to the requirements of the development goals, which serve as an essential parameter for the dynamic land use model to simulate the changes. These discriminating operations may not adequately represent specific planning scenarios, but to a certain extent, they outline the scenes of relevant planning scenarios.

Table 1.

Restriction and projection data of four policy scenarios.

Policy scenarios	Related goals	No-growth zone	Road network	Population in 2025 and 2035 (million)	Employed population in 2025 and 2035 (million)
S1 – Natural growth mode	None	Water areas	Current road network	98; 122	49; 64
S2 – Rapid economic growth mode	(3), (4)	Water areas, urban and rural development boundary	Current road network, planned major roads	102; 135	56; 78
S3 – Ecological and sustainable development mode	(1), (2)	Water areas, urban and rural development boundary, ecological reserves, agricultural land protection areas	Current road network	93; 118	48; 62
S4 – Comprehensive and coordinated development mode	(1), (2), (3), (4)	Water areas, urban and rural development boundary, ecological reserves, agricultural land protection areas	Current road network, planned tunnel and bridge across the Yangtze River	100; 125	52; 66

Note: The planned roads, tunnels and bridges are digitised from the 2017 Nanjing road network from OpenStreetMap (OSM).

Results

The future land use simulation results are first presented through a probability map of urban land use change (Figure S1 in the Supplemental material). The areas with a high probability of change are concentrated around the existing urban blocks and near the main traffic corridors. In contrast, the areas with a low probability of change are distributed around the areas with a high probability of change, showing outward diffusion. We need to make future urban growth areas meet urban development goals and expectations. Thus, we extracted specific targets related to land use change from Nanjing’s Master Plan for 2035 documentation (Table S2 in the Supplemental material) as the basis of change probability thresholding.

We normalized the value of possibilities to between 0 and 1 and divided them into 200 quantiles to produce comparable results. Then, we verified the area of land use demand in each scenario in ArcGIS and calculated the maximum and minimum values of change possibilities to meet the target conditions in all scenarios except the natural growth mode. For the final possibility thresholding, we took the averages of the average change probability of each of the three scenarios as the threshold that satisfies the condition. Thus, we generated the land use change results of the four pre-set scenarios with the same change possibility threshold (Figure S2 in the Supplemental material). From the overall development trend of the four scenarios, the results are consistent with the main development direction of the city, including the development along the banks of the Yangtze River, the outward diffusion of multiple urban blocks as the centre and the connection among city sub-districts. Although the results show similarity in the overall changed areas, differences in details exist.

The changed area and change direction of urban land use are summarised to analyse the differences in alternative land use types in spatial changes under the four scenarios (Table S3 in the Supplemental material). First, the total urban built-up area in 2035 shows that all the scenarios except the natural growth meet the development goal in the planning document (total size of urban land less than 2150 km²). Second, in the case of no-restricted-growth areas, the conversion of this land to urban land more or less occupies the ecological preservation area (72.36 and 55.75 km² for natural and rapid economic growth, respectively). Third, the difference in the changed area in the four scenarios also reflects the difference in the urban population and employment population projections, namely, the different comprehensive demands for urban land.

In addition, we exported the spatial overlay diagrams and highlighted three major areas in Figure 4 to study the differences in urban land use change at the following locations: (A) Luhe, (B) Jiangning and (C) Lishui. These three areas represent the fringe areas of the central city, the rapidly rising new city centre and the suburb, respectively. The prediction results under the four scenarios indicate that the concentration of unused and unbuilt land will become sensitive areas of land use change in the next 16 years. For the natural growth mode, the urban land expansion occurs around the edge of the central city and expands the urban land in the far centre because this scenario does not have strict limitations about the growth area and only follows the natural growth projection. Under the rapid economic growth mode, the change of urban land use is mainly reflected in the expansion of the main urban area to the suburbs, and the urban land in the north side of the Yangtze River increases significantly. Moreover, the suburban development is restricted with less expansion compared to the natural growth mode because this scenario takes into account the restriction boundary of urban and rural development, which is also a means for local governments to ensure the land development intensity and regulate the transformation of land use types. Under the ecological and sustainable development mode, the urban construction land expands in an orderly manner outside of the ecological protection and restricted construction areas because of the restriction imposed by green protection zones (i.e. no land use change in the defined agricultural land areas). Lastly, the estimated change in urban land would be relatively uniform in both central and suburban areas because the comprehensive and coordinated development mode encourages the urban land growth in an orderly and rational manner while meeting the needs of urban areas in different regions. The excessive expansion of urban land caused by the rapid economic growth will not occur because of the combination of the former three scenarios. The ecological and agricultural land areas are not affected.

Figure 4.

Growth scenarios for Nanjing in 2035: (a) NG (b) REG (c) ESD (d) CCD.

Discussion

Through the case study, four pre-set policy scenarios were predicted using the scenario-based spatial dynamic modelling method. The simulated land use changes present differences in area, location and volume under different combinations of planning and development strategies. We reveal and summarize the policy scenario implications to explain the difference in these scenarios as follows. First, rapid economic development may negatively impact the urban ecosystem, but the degree of impact can be mitigated by policies such as location and construction restrictions. However, the cost of preserving the ecosystem entails a high cost on urban land use development. Second, the scenario with an environmental protection focus indicates that land use development in Nanjing can meet the needs of future residential and business activities without encroaching on ecological and agricultural land areas. Without environmental protection policies, urban land use is very likely to occupy high ecosystem service areas because these areas are attractive to high-value commercial developments (e.g. areas accessible to natural resources and water) and residential developments (e.g. areas with highly scenic landscapes and natural reserves). Third, the ecological impact assessment indicates that the valuable ecosystem on the urban fringe is more susceptible to developmental pressures, specifically when considering the continuing urban expansion to the edges, which is so prevalent in rapidly developing Chinese cities at present.

Overall, this case study has shown that the modelling-based ‘what if’ scenario analysis has a positive reference that is significant in supporting the decision-making process of urban land use planning (Du and Li, 2005; Murtha et al., 2019). The spatial dynamic modelling simulates the future form and distribution of urban land use change on the city and regional levels by combining the dynamic land use model with scenario building. With appropriate adjustments to the elastic parameters in the model, different planning scenarios can be designed and tested. Different from previous studies (Pan et al., 2018; Zheng et al., 2012), the model contains a sub-module of information transformation of policy and planning strategies, which provide more intuitive information to help decision-makers better weigh the benefits and trade-offs among different scenarios. On this basis, the ‘impossible’ growth forms at the beginning of planning were filtered out. Proper and optimised scenario parameters were subsequently identified by the dynamic land use model and used as supplementary information for decision-makers. The quantitative results communicated in a visual form improve the practicability of the research output. Moreover, the modular implementation of different land use change scenarios for research is based on a summary of the services provided by the relevant impact factors, thus helping to integrate empirical knowledge and computation-based algorithms. All these have proven that the established scenario-based spatial dynamic model is a transparent and cohesive manner of planning support description and demonstration.

Conclusions

Desired future urban development can be realized by setting goals and formulating policies, while the phased development indicators affected by policies need specific means to demonstrate their feasibility. This study evaluated the effectiveness of planning policies by establishing a spatial dynamic model integrated scenario planning approach, which assists policymakers and increases the interaction of stakeholders in the decision-making process. Essentially in the case study of cities, the land use planning of a city, as the carrier of urban development, is often confronted with the interwoven influence of national, regional and urban policies. A spatial dynamic model with policy interpretation and transformation helps to coordinate and evaluate these policies at all levels, which also makes up for the lack of policy considerations and corresponding problems in the modelling process of the numerical input of driving variables. Therefore, the integrated use of this modelling approach should be advocated when dealing with complex planning policy and management issues at multiple scales.

Methodologically, this study presents a novel approach that integrates a scenario-based spatial dynamic modelling method with the information module of planning policy to better predict urban land expansion. This integrated approach takes into account both the socioeconomic and environmental driving factors of land use change and the possibly adjustable constraint conditions and projections of driving data to facilitate the scenario analysis of policy-driven ‘what if’ questions. The specific contributions can be summarised as follows:

The integration method of policy strategy analysis and spatial dynamic modelling provides a visual representation of various planning scenarios that helps us understand the potential impact of the policy decisions.

The comprehensive dynamic land use model simulates the future urban development form and patterns, urban capacity and its development potentials, which provides an essential basis for evaluating and optimising the urban planning scenario.

The transition from manual planning tasks to model input parameters illustrates the trade-offs associated with complex problems that help generate public feedback with more realistic expectations.

In conclusion, this study sheds light on combining a spatial dynamic modelling PSS with urban scenario planning. The means of policy scenario simulation provides additional visual and quantitative information as the basis for planning policy evaluation. However, method optimisation in terms of simulation accuracy and modelling synthesis remains a challenging task. For future research, we suggest continued improvement of the scenario-based spatial dynamic model, such as additional transport impact modules and more quantitative policy information extraction. Moreover, establishing a stakeholder-guided spatial change probability threshold mechanism will bring more diversified solutions to the growth scenario of realistic planning scenarios.

Supplemental Material

sj-pdf-1-epb-10.1177_2399808320934818 - Supplemental material for Spatial dynamic modelling for urban scenario planning: A case study of Nanjing, China

Supplemental material, sj-pdf-1-epb-10.1177_2399808320934818 for Spatial dynamic modelling for urban scenario planning: A case study of Nanjing, China by Zipan Cai, Bo Wang, Cong Cong and Vladimir Cvetkovic in Environment and Planning B: Urban Analytics and City Science

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the National Natural Science Foundation of China (41901191) and the Fundamental Research Funds for the Central Universities (19lgpy42).

Zipan Cai is a PhD candidate at the Department of Sustainable Development, Environmental Science and Engineering, KTH Royal Institute of Technology. His research includes urban spatial planning, spatial dynamic modelling, planning support system, and smart city studies.

Bo Wang is an Associate Professor at the Department of Urban and Regional Planning, School of Geography and Planning, Sun Yat-Sen University. His research interest includes urban and regional planning, ICT and travel behaviour, and smart city development.

Cong Cong is a PhD candidate at the Department of Urban and Regional Planning, University of Illinois at Urbana-Champaign. Her research focuses on urban transportation and land-use models.

Vladimir Cvetkovic is a Professor at the Department of Sustainable Development, Environmental Science and Engineering, KTH Royal Institute of Technology. His research has been mostly focused on water resources involving theoretical and experimental studies. Since 2015, he has been coordinating and working on projects related to urban systems in particular the socio-technical developments towards sustainable smart cities.

ORCID iDs

Zipan Cai

Bo Wang

Supplemental material

Supplemental material for this article is available online.

References

Amer

Daim

Jetter

(2013) A review of scenario planning. Futures 46: 23–40.

Barredo

Demicheli

Lavalle

, et al. (2004) Modelling future urban scenarios in developing countries: An application case study in Lagos. Environment and Planning B: Planning and Design 32(1): 65–84.

Berg

Rogers

Mineau

(2016) Building scenarios for ecosystem services tools: Developing a methodology for efficient engagement with expert stakeholders. Futures 81: 68–80.

Bezold

(2009) Aspirational futures. Journal of Futures Studies 13(4): 81–90.

Burian

Stastny

Brus

, et al. (2015) Urban planner: Model for optimal land use scenario modelling. Geografie 120(3): 330–353.

Carr

Zwick

(2005) Using GIS suitability analysis to identify potential future land use conflicts in North Central Florida. Journal of Conservation Planning 1: 89–105.

Carr

Zwick

(2007) Smart Land-Use Analysis: The LUCIS Model. Redlands: ESRI Press.

Chakraborty

Kaza

Knaap

, et al. (2011) Robust plans and contingent plans: Scenario planning for an uncertain world. Journal of the American Planning Association 77(3): 251–266.

Chen

Liu

, et al. (2020) Global projections of future urban land expansion under shared socioeconomic pathways. Nature Communication 11: 537.

10.

Deal

Pan

(2017) Discerning and addressing environmental failures in policy scenarios using planning support system (PSS) technologies. Sustainability 9(1): 13.

11.

Deal

Pan

Timm

, et al. (2017) The role of multidirectional temporal analysis in scenario planning exercises and planning support systems. Computers, Environment and Urban Systems 64: 91–102.

12.

Deal

Schunk

(2004) Spatial dynamic modeling and urban land use transformation: A simulation approach to assessing the costs of urban sprawl. Ecological Economics 51(1–2): 79–95.

13.

Dendoncker

Rounsevell

Bogaert

(2007) Spatial analysis and modeling of land use distributions in Belgium. Computers, Environment and Urban Systems 31(2): 188–205.

14.

(2005) Planning support system (PSS) and its application to decision-making for urban spatial development. Journal of Wuhan University (Engineering) 38(1): 137–142.

15.

Foley

Defries

Asner

, et al. (2005) Global consequences of land use. Science 309(5734): 570–574.

16.

Geertman

(2002) Participatory planning and GIS: A PSS to bridge the gap. Environment and Planning B: Planning and Design 29(1): 21–35.

17.

Geertman

Stillwell

(2009) Planning Support Systems Best Practice and New Methods. Dordrecht: Springer.

18.

Geertman

Stillwell

(2012) Planning Support Systems in Practice. Berlin: Springer.

19.

Inayatullah

(2008) Six pillars: Futures thinking for transforming. Foresight 10(1): 4–21.

20.

Johnson

Sieber

(2009) Agent-based modelling: A dynamic scenario planning approach to tourism PSS. In: Geertman

Stillwell

(eds) Planning Support Systems Best Practice and New Methods. Dordrecht: Springer, pp.211–226.

21.

Kaiser

Godschalk

Chapin

(1995) Urban Land Use Planning. Chicago: University of Illinois Press.

22.

Klosterman

(1999) The what-if? Collaborative planning support system. Environment and Planning B: Planning and Design 26(3): 393–408.

23.

Klosterman

Pettit

(2005) An update on planning support system. Environment and Planning B: Planning and Design 32: 447–484.

24.

Koontz

(2005) We finished the plan, so now what? Impacts of collaborative stakeholder participation on land use policy. Policy Studies Journal 33: 459–481.

25.

Kosow

Gaßner

(2008) Methods of future and scenario analysis: Overview, assessment, and selection criteria. Deutsches Institut Fur Entwicklungspoliti Studies 39: 133.

26.

Liu

Silva

(2017) Examining the dynamics of the interaction between the development of creative industries and urban spatial structure by agent-based modelling: A case study of Nanjing, China. Urban Studies 55(5): 1013–1032.

27.

Liu

Liang

, et al. (2017) A future land use simulation model (FLUS) for simulating multiple land use scenarios by coupling human and natural effects. Landscape and Urban Planning 168: 94–116.

28.

Liu

Fang

(2014) Key issues of land use in China and implications for policy making. Land Use Policy 40: 6–12.

29.

Yang

(2002) Drivers of land-use/land-cover changes and dynamic modeling for Atlanta, Georgia Metropolitan Area. Photogrammetric Engineering and Remote Sensing 68(10): 1073–1082.

30.

Long

Shen

Mao

(2012) Retrieving spatial policy parameters from an alternative plan using constrained cellular automata and regionalized sensitivity analysis. Environment and Planning B: Planning and Design 39(3): 586–605.

31.

Long

Zhang

(2015) Land-use pattern scenario analysis using planner agents. Environment and Planning B: Planning and Design 42: 615–637.

32.

Loo

BPY

Wang

(2018) Factors associated with home-based e-working and e-shopping in Nanjing, China. Transportation 45(2): 365–384.

33.

Meissner

Wulf

(2013) Cognitive benefits of scenario planning: Its impact on biases and decision quality. Technological Forecasting and Social Change 80(4): 801–814.

34.

Mitsuda

Ito

(2011) A review of spatial-explicit factors determining spatial distribution of land use/land-use change. Landscape and Ecological Engineering 7(1): 117–125.

35.

Murtha

Wang

Duan

(2019) Historical analysis of land use change and geo-design of rapid urbanization: Orlando, Florida, USA. Journal of Digital Landscape Architecture 4: 205–212.

36.

Pan

Deal

(2019) Reporting on the performance and usability of planning support systems—Towards a common understanding. Applied Spatial Analysis and Policy 13: 137–159.

37.

Pan

Deal

Destouni

, et al. (2018) Sociohydrology modeling for complex urban environments in support of integrated land and water resource management practices. Land Degradation & Development 29(10): 3639–3652.

38.

Pan

Page

Zhang

, et al. (2019a) Using comparative socio-ecological modeling to support climate action planning (CAP). Journal of Cleaner Production 232: 30–42.

39.

Pan

Zhang

Cong

, et al. (2019b) A dynamic and spatially explicit modeling approach to identify the ecosystem service implications of complex urban systems interactions. Ecological Indicators 102: 426–436.

40.

Perveen

Kamruzzaman

Yigitcanlar

(2017) Developing policy scenarios for sustainable urban growth management: A Delphi approach. Sustainability 9(10): 1787.

41.

Raford

(2015) Online foresight platforms: Evidence for their impact on scenario planning & strategic foresight. Technological Forecasting and Social Change 97: 65–76.

42.

Schwartz

(2012) The Art of the Long View: Planning for the Future in an Uncertain World. New York: Crown Business.

43.

Shoemaker

PJH

(1995) Scenario planning: A tool for strategic thinking. Sloan Management Review 36(2): 25–40.

44.

Stojanovic

Mitkovic

(2014) The scenario method in urban planning. Facta Universitatis - Series: Architecture and Civil Engineering 12(1): 81–85.

45.

Tevis

(2010) Creating the future: Goal-oriented scenario planning. Futures 42(4): 337–344.

46.

Veldkamp

Fresco

(1996) CLUE: A conceptual model to study the conversion of land use and its effects. Ecological Modelling 85(2–3): 253–270.

47.

Volkery

Ribeiro

(2009) Scenario planning in public policy: Understanding use, impacts and the role of institutional context factors. Technological Forecasting and Social Change 76(9): 1198–1207.

48.

Wilson

(1994) Strategic planning isn’t dead – It changed. Long Range Planning 27(4): 12–24.

49.

(2015) Planning for Growth: Urban and Regional Planning in China. New York: Routledge.

50.

Xiang

Clarke

(2003) The use of scenarios in land-use planning. Environment and Planning B: Planning and Design 30: 885–909.

51.

Haase

, et al. (2020) Assessment of landscape changes under different urban dynamics based on a multiple-scenario modeling approach. Environment and Planning B: Urban Analytics and City Science. DOI: 10.1177/2399808320910161.

52.

Zang

, et al. (2011) Analyzing and modeling land use land cover change (LUCC) in the Daqing city, China. Applied Geography 31: 600–608.

53.

Zegras

Rayle

(2012) Testing the rhetoric: An approach to assess scenario planning’s role as a catalyst for urban policy integration. Futures 44(4): 303–318.

54.

Zhao

Huang

Zhang

(2014) Ecological risk assessment of provincial land-use overall planning in China. Human and Ecological Risk Assessment: An International Journal 20(6): 1491–1506.

55.

Zheng

Zhao

Xiang

, et al. (2012) A coupled model for simulating spatio-temporal dynamics of land-use change: A case study in Changqing, Jinan, China. Landscape and Urban Planning 106(1): 51–61.

56.

Zhou

Huang

Chen

, et al. (2017) The effect of land use planning (2006-2020) on construction land growth in China. Cities 68: 37–47.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.50 MB