A Bayesian approach to complex land use change modelling

Lead Research Organisation: University of Surrey
Department Name: Civil and Environmental Engineering


The UK is currently facing a housing shortage and the government plan is to increase house building to a rate of 300,000 per year in the mid-2020s. This is largely supported through policy instruments related to improving access to personal finance and reducing planning barriers. According to the government, over 40 per cent of local planning authorities do not have a plan that meets the projected growth in households in their area. Against this backdrop of announced substantial, poorly understood and more liberally regulated land use change, there is a clear need to understand possible patterns of future urban growth.

There exist computational models that take economic and demographic forecasts and use these to explore possible trajectories of land use change. A particular family of such models are Cellular Automata (CA) land use models; these models are complex in a mathematical sense as they represent the chaotic and spatially uneven processes by which cities and towns can grow. A major hurdle preventing the widespread use of these models is that the better performing models are calibrated based on 'expert judgement'. This reliance on manual intervention is not only resource intensive but also casts a shadow on the objective reliability of model results.

In this project we will investigate the use of a statistical technique of Approximate Bayesian Computation to automatically calibrate CA land use models. We aim to deliver a proof-of-concept demonstrating the methods on a CA model that is stripped down to its fundamental essentials, and applied to a variety of UK and European urban areas. A further advantage of the Bayesian approach is that it will not just give the best-fitting set of parameters, but also the associated uncertainty. This will allow analysts to use the model to arrive to support decisions with confidence based on a range of possible outcomes and their likelihood.

Planned Impact

The project will allow the characteristics of urban growth across a number of European functional urban areas to be compared by estimating land use change models. This is of immediate policy relevance as it provides insight into how patterns of change could have been different and what future trajectories are possible, in addition to the historical trends. The UK is currently facing a housing shortage and the government plan is to increase house building to a rate of 300,000 per year by the mid-2020s. This is largely supported through policy instruments related to relaxing spatial planning restrictions and improving access to personal finance. According to the government over 40% of local planning authorities do not have a plan that meets the projected growth in households in their area. Consequently, it is anticipated that local authorities would be interested in tools that help foresee and manage urban growth, against this backdrop of announced substantial, poorly understood and more liberally regulated land use change.

The project aims to deliver a proof-of-concept, and further research and development will be necessary to deliver that impact to its fullest. Our actions supporting the pathways to impact are directed at facilitating uptake in research and development. The project can have a positive impact on the future direction of land use change research by demonstrating an overall approach to evidence based complex land use change modelling, and by making the composite elements of that approach widely available in an accessible form. This will be facilitated by sharing the source code and scripts of the full procedure in open source on the Github website which is the globally leading website for open source code sharing. It can reach professional users beyond academia, including analysts and consultants in the area of spatial planning and policy. By sharing the project outcomes in this manner, the foundation and platform for further dissemination will be established
The expected impact outside of research and development is through the application of the methods in existing or new planning support systems (PSS). As the project is aiming for an integrated solution, there is also the possibility of developing the method into a new product to be managed by a commercial partner as the project does present a unique capability of giving objective insight in the range of possible trajectories of future land use change. We will under the current bid develop communication material (PowerPoint slides, brochure, poster) highlighting the potential application of the developed methods rather than the underlying technical innovation. This will be directed at existing partners active in land management, such as local authorities, and national infrastructure operators and facilitate a discussion with them about possible future directions and applications of this work.
Description The project developed a novel statistically robust approach for forecasting future growth trajectories of towns and cities, in particular the development of the urban rural interview. A number of technical innovations have been made, and there is also a new practical application. Existing models of urban growth have focused on aimed to extrapolate existing growth patterns and projecting the most likely future scenario given economic and demographic constraints. In the current project we developed a method of identifying modes of growth, allowing us to provide growth projections under different planning regimes. This approach is of interest to spatial planners because it gives them better insight in the decision space and context available to them.
Exploitation Route Yes, the work is relevant in different areas. The work demonstrates an effective integration of processes, modelling, landscape characterisation and Bayesian estimation that is relevant to the wider field of spatio-temporal analysis. Researchers interested in urbanisation can use the models and model results as a tool of studying urbanisation patterns and its consequences. For these researcher the statistical robustness, and its treatment of uncertainty in predictions and future developments is especially relevant. In our own network we have seen interest in application in relation to ecosystem service analysis. The analysis of growth modes offers perspective for international comparative analysis.
Sectors Communities and Social Services/Policy,Construction,Digital/Communication/Information Technologies (including Software),Environment,Government, Democracy and Justice,Transport

URL https://doi.org/10.1016/j.compenvurbsys.2021.101689
Title Urban growth simulation model 
Description This software is a simulation tool for creating trajectories of the spatial development of future urban growth trajectories. The software incorporates the major innovations made in the project with respect to model formulation, model estimation and model diagnostics. The software is a library and includes a demonstrative application that replicates all of the results of the associated journal paper. Yu, J., Hagen-Zanker, A., Santitissadeekorn, N., & Hughes, S. (2021). Calibration of cellular automata urban growth models from urban genesis onwards-a novel application of Markov chain Monte Carlo approximate Bayesian computation. Computers, environment and urban systems, 90, 101689. 
Type Of Technology Software 
Year Produced 2021 
Open Source License? Yes  
Impact The software has excellent potential for the analysis of urban dynamics across geographical areas and time periods, and can be used with existing global datasets (in particular the Global Human Settlement Layer). This wider application is currently being demonstrated by the by the project team in a wider pan-European application 
Description Presentation at Advanced Urban Modelling seminar 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact This was a presentation in a seminar series on Advanced Urban Modelling. Normally, this would reach the usual audience of our peer network. This year it was live-streamed and made widely available. There was a large number of participants that were undergraduate or postgraduate students, and also there were more participants / audience members with a professional rather than research background.
Year(s) Of Engagement Activity 2020
URL https://www.martincentre.arct.cam.ac.uk/conferences/AUM/aum-2020/session-6-effects-of-the-covid-19-p...
Description Presentation at Landscape Decision seminar (20 May) disseminated on YouTube (https://www.youtube.com/watch?v=u7JdpYJOBnM&list=PLrlZ6FipN5mlS4_lN9PwZAR6HCyppGTDP&index=2&t=15s) 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact This was part of a wider effort of communication and networking of the Landscape Decisions project. The primary impact was better awareness across the programme of parallel activities.
Year(s) Of Engagement Activity 2020
URL https://www.youtube.com/watch?v=u7JdpYJOBnM&list=PLrlZ6FipN5mlS4_lN9PwZAR6HCyppGTDP&index=3&t=15s