A regional climate model (RCM) at 25 km resolution, coupled to a simple urban land-surface scheme has been used to assess the sensitivity of UK urban climates to large-scale greenhouse gas (GHG) induced climate change, local forcing from urban land use, and anthropogenic heat flux (AHF) resulting from energy use in urban areas. GHG-induced climate change is similar over both urban and rural land surfaces in the model, suggesting that under a changing climate the relative magnitude of urban heat islands in the UK would remain the same. However, indirect effects from urbanisation or changes in energy consumption within UK cities would contribute to further localized anthropogenic climate change that can be quantified by this model. The presence of the urban heat island also has a significant impact on the frequency of extreme high temperatures both in present and future climates, and will exacerbate the impact of climate change on the urban population.The simple urban land-surface exchange scheme is a cost effective additional component of the RCM. It is shown to provide a realistic city-scale urban heat island responding to the dynamical weather of the model at every model time-step, with a magnitude and variability comparable to observed values. The climate model provides long integrations to derive large samples of heat islands under a full range of simulated dynamical conditions, and in this study sampling more than 100 years of climate change.

In this paper, we estimate the climatological impact of urbanisation in the UK as derived from the present and historical set of temperature observations from the UK network of meteorological observing stations. A well-established method for interpolation of in-situ temperature observations is used to make an estimate of temperatures at urban weather station locations based on rural climate data. This method avoids reliance on simple pairing of urban–rural stations commonly used for station-based analysis, and explicitly accounts for other topographic features such as elevation and coastal effects that will introduce geographical differences in temperature. The method is then applied to select a network of ‘rural’ weather stations deemed to be largely free from the direct influence of urban micro-climate effects. Comparison of the temperature grids derived from the ‘urban’ and ‘rural only’ station sets enables quantification of climatological urban heat island intensity across the UK and the influence on temperature in the UK national observational network observations from urban stations. Practical application: This new method allows the estimation of the urban heat island for towns and cities of any size. Researchers will find it useful when making predictions of the urban heat island and of climate change in the urban environment, and when investigating morbidity and mortality due to overheating in buildings.