Author:

Abstract:

More than half of the world population is at present living in cities. Especially the least developed countries are playing a dominant role in urbanization. The rapid pace of this growth is posing the rather weak urban governments to severe challenges in providing the necessary employment opportunities, services and infrastructure required to comfortably host the growing urban population. Moreover, whereas urban areas are considered as places of concentration of livelihood opportunities and economic welfare, not all social groups have access to this wealth. Income inequality is growing and spatially translated in social segregation whereby socially weak population groups fear the risk to become trapped in urban poverty. The overall objective of this thesis is to better understand urban growth and social segregation by integrating (1) urban growth modelling with (2) urban livelihood studies. Traditionally, urban growth modellers adopt quantitative GIS-based models, while livelihood studies are mostly approached via qualitative research methods. On their own both approaches show severe limitations. The hypothesis of this thesis is that bridging the gap between both approaches allows to address new questions regarding the impacts of urban growth on urban livelihoods and vice versa, which are highly relevant for the development of sustainable urban policy strategies. In developing countries there is an urgent need for uncomplicated and straightforward spatially explicit models that are capable of generating expansion scenarios in order to support spatial planning decision making. Therefore, a logistic regression model based on freely available satellite data allowed to identify the spatial growth trend of our case study Kampala (Uganda) which identified the road pattern, the city centre, topographic setting and existing built-up as important spatial driving factors of urban growth. Combined with available census data this model is used to simulate alternative growth scenarios that allow to evaluate potential policy strategies such as zoning, densification, and the development of urban infrastructure and satellite cities. The effect of such future urban growth scenarios on livelihood strategies is illustrated through quantifying the potential land losses of different types of urban. Three types of urban farmers are identified: subsistence farmers, garden farmers and commercial farmers. According to an urban sprawl scenario, subsistence farmers are assessed to loose spatially many opportunities while infilling is rather targeting garden and commercial farmers as spatial victims. Such analyses allow to evaluate urban trends and policy strategies. Next, research revealed that in order to understand social segregation, insights in urban livelihood diversity is necessary. Since mobility plays an important role in daily lives, urban livelihood is studied within the scope of urban mobility. First, major mobility constraints of different social groups are identified through participant observation which identify potential governmental action in order to guide sustainable urban development. Further, a livelihood typology of all interviewed residents is carried out based on socio-economic characteristics which are then linked to their mobility pattern. The typology confirmed that income forms the main criterion in urban areas to distinguish residents: we found extreme poor, poor, middle income and rich residents as the four main livelihood groups in Kampala. A further observation and inquiry of the daily travel to work amongst those groups provide insights which allowed to evaluate potential policy strategies such as the planned bus rapid transit system of Kampala’s planning authorities. Results show that BRT might deliver a time gain of on average 11 minutes and that especially extreme poor travellers have the highest potential to benefit from the system. However, besides time gain, financial constraints must be considered which according to present plans indicate the BRT to be unaffordable to these extreme poor. Finally the results of the livelihood typology are used to identify residential preferences of the different resident groups which are used to develop ASSURE, an agent-based model that allows to simulate urban growth and social segregation. The model translates residential habits into utility functions which allow agents to rate different locations in the city in terms of accessibility, quality of living and affordability. Also, interaction rules are introduced which allow higher social classes to force lower social classes to relocate. These dynamics are used to simulate future growth scenarios and social segregation according to considered policy strategies (e.g. zoning, densification and satellite city development) and socio-economic trends (e.g. wealth increase). The results are explorative and rely on many assumptions, but nevertheless the model revealed high potential as decision support tool to urban policies through exploring potential impacts of specific policies.