Computational modelling of fires in the slums of Mukuru, Nairobi

Informal settlements (slums, favelas, shanty towns etc.) are already home to 1 billion people globally, and are likely to grow extensively in the future as it is estimated there will be further urban population growth of 2.5 billion people in the next three decades. Informal settlements are often highly vulnerable to settlement-wide fires which spread quickly and extensively due to the densely packed space and light building materials. The rate at which these settlements are currently growing necessitates their integration with the wider, formal urban landscape, rather than their removal. Part of this work must include the finding of solutions to the problem of destructive fires, which entrench already deep poverty cycles.

The proposed research will investigate informal settlement fires, their development and propagation, and possible mitigating measures through the use of computer-based modelling techniques (computational fluid dynamics and cellular automata). It will primarily focus on urban environments with settlement dwelling characteristics found in the Mukuru slum of Nairobi, Kenya. The work may also include experimental testing to determine the thermal and combustion properties of the building materials prevalent in these settlements. This is required to inform the selection of accurate input parameters to the computer models. The main objectives will be to model fire growth and development in informal dwellings and settlements as they currently exist, and to then compare the results against further models which incorporate theoretical prevention and mitigation measures. These measures will be conceptualised with the understanding that it is difficult to apply large scale alterations in the physically challenging informal environment. Computer-based modelling provides the benefit of being able to run tests that would be too large and/or expensive as real-scale experiments, as well as providing easier repeatability for investigating the variation of physical parameters.

This work is a development on CFD modelling that has already been conducted as part of the IRIS-Fire project which has looked at informal dwellings in South Africa. Informal dwellings found in Nairobi are of particular interest in comparison to the informal dwellings of South Africa as they have roofs that are pitched rather than flat, and are often connected in long rows rather than being standalone structures. Furthermore, some even extend over two storeys, providing another element of fire risk which is not necessarily prevalent in other informal environments. It is important to understand the similarities and differences in fire behaviour across different informal environments, to discern between development strategies that will be effective universally, and those that are only relevant in certain locales. This work should be of benefit to NGOs and local authorities (particularly in Nairobi) that are concerned with development of informal settlements. It may also be of relevance to the work of the School of Geoscience's multi-hazard urban disaster risk hub.