Effects of urban areas on the diurnal cycle of temperature and precipitation as found in global climate simulations

Urban areas with their high population density, industrial facilities and traffic account for about 70% of the world’s CO₂ emissions and thereby contribute to global climate change. In addition, the urban fabric and its qualities (albedo, heat storage, vertical extension) influence atmospheric parameters, resulting in the well-known urban heat island (UHI) effect that may influence precipitation as well as temperature. Although both effects are found worldwide, analyses of global climate model results for urban impacts on temperature and precipitation are rare.

Previous analyses of global climate model CCAM simulations for 1985-2010 have revealed UHI effects and influences of urban areas on minimum and maximum temperatures [1]. Using the same data set to derive time-zone corrected 3-hourly data, climate average global diurnal cycles of temperature and precipitation were calculated. The  UHI intensity depends on season and climate region (northern extra-tropics (NET), southern extra-tropics (SET), or tropics (TR)). The largest statistically significant UHI influences are consistently found in the evening and at night-time, in agreement with many previous studies. Signs of urban cool islands were found for a few hours per season and climate region. Influences of urban areas on precipitation varied, with increases and decreases in all climate regions and seasons. For NET (274 cities), two out of three cities show significantly increased precipitation, while for SET (39 cities) and TR (26 cities) more decreases than increases were noted. Exceptions to these general influences of urban areas on precipitation were found at several hours for all seasons. The geographic settings of the large urban areas used for these analyses might be one explanation of the different precipitation changes induced by UHI, since urban areas are not distributed evenly around the globe. The increase in precipitation seen for NET might be because these urban areas are in less warm climate regions compared to SET and TR.

[1] Katzfey J.; Schlünzen K.H.; Hoffmann P.; Thatcher M. (2020): How an Urban Parameterization Affects a High-resolution Global Climate Simulation. QJRMS, 146 (733), 3808–3829. https://doi.org/10.1002/qj.3874.