How will the urban heat island in growing cities behave under climate change?

The UHI is caused primarily by the absorption of heat by buildings, the trapping of that heat in enclosed spaces, and the reduced evaporation from built areas compared to natural vegetation. Further heat sources include heat from tarmac and heat emissions by buildings in winter. The UHI has been documented for a number of European (Jones and Lister, 2009), American, and Chinese cities using long time series of paired data (urban/rural) covering multiple-decades to century, or using local spatial variations in temperature based on satellite observations (Zhou et al., 2013).

The UHI is generally characterised at the local-to-regional level because of the specific data requirements and regional interests. Published studies indicate that it grows initially as the city develops, but it tends to saturate when urbanization reaches a certain level of maturity. However, these studies cover a limited number of cities and geographical/climate settings. Thus we do not fully understand what factors determine the rate of growth and the saturation level of the UHI and the extent to which these factors may have a different influence in say coastal or arid settings.

A systematic assessment of UHI warming will be undertaken, initially based on all available long time-series paired station data (i.e. including data within and outside cities). The availability of high-quality rural-city station pairs is however limited both for individual cities and geographically. Thus it is also proposed to use Land Surface Temperature (LST) data, for example from the MODIS spectroradiometer. LST is not directly comparable in absolute terms to 2m air temperature measurements but can be used to explore spatial patterns and to help in the interpretation of the limited number of longer station records.

By combining this UHI data with data on the environmental and socio-economic characteristics of cities (such as size, land use, climate regime, geographic setting, green/blue space, albedo) mechanistic relationships will be identified between the characteristics of the UHI and the underlying drivers.

Outputs from high-resolution global (CMIP5 and CMIP6) and regional (CORDEX) climate models will also be used. Apart from some limited sensitivity studies (McCarthy et al., 2012) such models do not implicitly model the UHI, so, if time permits, the possibility of using a relatively simple scaling approach in which temperature change statistics from the models are used to modify the present-day urban climate statistics will be explored. This would allow the development of climate projections which do account for the UHI and thus are more appropriate for impact studies.

The project will also explore the implications of projected UHI effects this century for heat-related health burden (using heat stress related indices such as Wet Bulb Globe Temperature), and possibly propose solutions to minimise it. Although the project will focus on heat, some attention will also be given to humidity as it has important implications for health and human comfort (and the extent of evaporative cooling helps to determine the magnitude of the UHI).