A global perspective on the urban heat island effect

The areal coverage of urban areas is expanding in many parts of the globe, with 70% of the world's population projected to live in towns and cities by 2050.The urban heat island effect can cause cities to be 6-12C warmer than the surrounding rural areas, which is of great concern due to the increased fatalities associated with heatwaves, an issue that will be exacerbated by climate change.
As in many models used for global weather and climate prediction, urban areas are currently absent from the model of the European Centre for Medium-Range Weather Forecasts (ECMWF),but they are in the process of being introduced with a view to becoming operational at 9km resolution in the next few years.This involves modification of the surface roughness and water availability, and a rigorous treatment of the turbulent and radiative exchanges at the principal elements of an urban area (the streets, walls and roofs).This PhD project is an excellent opportunity to be involved in this exciting work,from development of the underpinning science of modelling urban areas, through to the implementation and exploitation of this new capability in the ECMWF system to answer important scientific questions on a global scale.
The proposed approach to the fast treatment of 3D radiative exchanges in urban areas is at the forefront of the science of radiative transfer modelling but will require several additional features to be developed and evaluated.A potential advantage will be that very few parameters will be required to describe the morphology of urban areas.An important task will be to examine existing global datasets to determine appropriate values to use for different urban forms globally.
With urban areas added in on a global scale, there will be the capacity to run a fully interactive global weather forecast model.This will enable s a large number of new questions about the impact of urban areas, and the importance of cities of different sizes, to be addressed more comprehensively than has been done to date.
The PhD project would have three components involving a number of key questions:
1.How can we optimally represent 3D radiative transfer in urban areas? A good starting point will be a new method "SPARTACUS" that has recently been developed for efficiently representing the 3D interaction of solar and thermal-infrared radiation with randomly distributed objects and has been applied successfully to both clouds and vegetation. The studentship will address what changes will be needed to adapt it for buildings? What is the impact on radiation of incorporating trees and parks into the urban landscape? How well does this approach match benchmark calculations from much more detailed (and computationally expensive) radiation models?
2.How well does the new urban treatment in the ECMWF model predict near-surface variables (e.g. air temperature and surface fluxes of heat and moisture)? A systematic evaluation will be performed against existing observations from a number of contrasting city areas from around the world, and possibly also using satellite observations of skin temperature. By re-running the global land-surface module (which is fast) of the ECMWF model,the student will investigate how improvements in the treatment of urban areas in the model can improve performance relative to observations.
3.Many intriguing scientific questions could then be asked using the full model. For example, What is the impact of urban areas globally on the weather? Whether or not the impact is significant is a topic of controversy in the literature, but the student will be able to exploit existing ECMWF software tools to determine whether forecasts are improved downstream of cities. What will be the impact of future increases in urban cover, and possible mitigation efforts such as building greener cities? By modifying the global maps of urban properties used by the model it will be possible to explore how the urban heat island effect could change in the future.