The cities of the Global South need to become more flood resilient in light of climate change and rapid urbanisation. In this thesis, a land-use modelling case study for the Mumbai Metropolitan Region is executed on the interaction of flood risk and land use, through which future flood resilience was assessed for several urban planning strategies. Concluded was that the urban environment will become less resilient in case no governmental intervention occurs, and urban planning can contribute to the resilience of the urban environment. Furthermore, the effectiveness of urban planning was largely dependent on the strictness with which these policies were implemented and regulated, but largely independent of the perceived climate future.

Urbanization makes cities more vulnerable in the face of climate change risks. Understanding urban growth under climate change can help planners optimize land allocation and development strategies that are resilient to the impacts of climate change. However, the complexity and uncertainty of climate change hinder the urban growth projections and make it hard to form local climate adaptation plans. Therefore, this research aims to develop a methodology to incorporate climate uncertainties into land-use models to explore plausible futures.

The literature review results indicate the lack of a methodology to quantitatively link climate change effects with land-use models and to systematically explore the full parameter space of the climate uncertainties. Hence, our main research question becomes “How can an integrated land-use modelling methodology be developed to help systematically explore the impacts of climate uncertainties on urban growth?” This research question is answered by integrating Metronamica, a cellular-automata-based land-use modelling framework with Exploratory Modelling. The Metropolitan Region of Amsterdam (MRA) is selected as the case city to demonstrate the proposed methodology, and the research scope focuses on flooding, a typical and important climate impact. Specifically, we include the flooding probability maps into the “suitability” section under Metronamica, based on the principle that the higher the flooding risk of an area, the lower the suitability value. These flooding suitability values are deemed as uncertain in our research and they are not given fixed values but certain uncertainty ranges.

The flooding factors and their defined uncertainty ranges are added to a model established for the case city. Then this model is connected with the Exploratory Modelling Analysis (EMA) workbench and generates 2000 experiments by the random samplings and combinations of the uncertainties. In the result analysis step, we use clustering algorithms to select 34 representative maps, followed by the comparisons between them and the base map where no flooding factor is included. The 34 representative maps show some land-use change characteristics because of the introduction of the flooding suitability variables and the uncertainties. These characteristics of the projected land-use outcomes are extracted and the reasons behind these observations are explored. Finally, the land-use changes from 2015 to 2050 with the flooding risk considerations are summarized. Based on the summarized changes, we formulate some policy implications for urban planning under climate uncertainties.

This research presents a complete workflow of dealing with climate uncertainties in land-use analysis, including the selection of methods, the model conceptualization to include climate variables, the model implementation to set up a Metronamica model for the case city, and the connection between the Metronamica and the Exploratory modelling techniques. In the last step, we carry out result analysis and interpretation, finding out the land-use changes caused by the inclusion of flooding into urban dynamics. This integrated framework could be further applied and improved in the intersections of climate change and urban development, to provide insights into climate adaptation and urban climate resilience.