The urban heat island

 

List of contents

Open Climate4you homepage

 

General

An urban heat island (UHI) is a metropolitan area which is significantly warmer than its surroundings. It is well known that urban areas are generally warmer than nearby rural areas. Often referred to as the urban heat island effect, urbanization has long been regarded as a potential serious contamination of the global climate signal (e.g., Landsberg 1956). Even for small settlements in Arctic environments, the temperature effect has been shown to be considerable (Hinkel et al. 2003).  

The main cause of the urban heat island effect is land surface changes, with roads and buildings taking up an increasing part of the total area, changing the overall reflectance (albedo) of the area. Surfaces covered by roads and buildings reduce evaporation, which is an important surface cooling process, thereby contributing importantly to higher air temperatures within urban areas. Heat generated by energy usage is a second contributor to UHI. As population centres grow they affect larger and larger areas, which usually will experience a corresponding increase in average temperature. Also the local precipitation pattern might be affected by large cities, especially in a downwind direction.

 

Major cities in Europe at night, seen from satellite.  

The degree to which UHI affects current estimates of global temperatures remains under debate. Peterson (2003), based on an analysis of urban and rural temperatures from 289 stations 1989-1991 in USA, concluded that the effect was small or negligible. On the other hand, a recent study by McKitrick and Michaels (2007) indicate that there is considerable influence of urban heat release on global temperature estimates. These authors conclude that removing the nonclimatic UHI effects would reduce the estimated 1980-2002 global average temperature increase over land by about half.

In conclusion, we still need to improve our understanding of urban heat islands, the local weather and climate within cities, and there might be a serious influence of this phenomenon on current global temperature estimates. The effect would be to make the 20th century global temperature increase larger than what is real. 

The existing network of meteorological stations was never designed as a means to measure the global surface temperature. Most of the meteorological stations was established to obtain information on weather where people are living. And that is precisely what they are doing. That is also why there is a regrettable lack of meteorological stations in the Arctic, the Antarctic, in high mountains, and elsewhere where the population density is low. Meteorological institutions are doing an excellent job and can definitely not be blamed for potential effects of UHI on global temperature estimates.  

Presumably the urban heat island phenomenon like other natural phenomena is complex, and requires detailed investigations to be fully understood. The effect on global temperature estimates should be carefully evaluated by well-designed measurements in and around urban sites. A more thorough investigation of the urban heat island phenomena probably requires the establishment of several measurement stations spread across the area investigated.

To exemplify the influence of UHI on surface air temperatures, a series of temperature traverses was carried out across Oslo ( Norway ), Longyearbyen ( Svalbard , Norway ), and St. Andrews ( Scotland ). The graphic result of these temperature measurements can be seen by clicking on the small diagrams in the tables below.

Please note that these preliminary investigations do not represent a snapshot of the temperature distribution, but were carried out over some time, from about 15 minutes to more than 2 hours. I addition, the temperature sensor were mounted on a moving vehicle, which itself could be a source of heat. This, however, applies to all measurements within each series, and there is little doubt that the overall features shown by the experiments are real.

Click here to jump back to the list of contents.

 

Urban heat island effects in Oslo (59o50'N, 10o40'E), southern Norway. Number of inhabitants: c. 550,000.

Year Day Traverse
2007 25 January
2007 7 June
2007 10 August
2007 3 September
2008 5 September
2009 3 July

Click on the individual small diagrams to open full-size diagrams.

Click here to jump back to the list of contents.

 

Urban heat island effects in Longyearbyen (78o17'N 11o20'E), Spitsbergen, Svalbard. Number of inhabitants: 2,001 (January 1, 2007).

Year Day Traverse
2007 2 February
2008 20 January
2008 26 January
2008 30 January
2008 31 January
2008 31 March
2008 18 April
2008 28 April
2008 17 July
2008 27 July
2009 24 April
2009 22 June
2010 2 May

Click on the individual small diagrams to open full-size diagrams.

Click here to jump back to the list of contents.

 

Urban heat island effects in St.Andrews (56o20'N 2o50'W), eastern Scotland. Number of inhabitants: c. 14,000.

Year Day Traverse
2007 13 December
2008 11 January
2008 25 February
2008 6 May
2008 11 June

Click on the individual small diagrams to open full-size diagrams.

 

Click here to jump back to the list of contents.

 

Other local meteorological phenomena

Year Day Phenomena
2008 1 August Effects of Svalbard solar eclipse
2008 2 September Recent Svalbard warming revisited

Click on the individual small diagrams to open full-size diagrams.

 

Click here to jump back to the list of contents.