Solar reflective urban surfaces (white rooftops and light-colored pavements) can increase the albedo of an urban area by about 0.1. Increasing the albedo of urban and human settlement areas can in turn decrease atmospheric temperature and could potentially offset some of the anticipated temperature increase caused by global warming. We have simulated the long-term (decadal to centennial) effect of increasing urban surface albedos using a spatially explicit global climate model of intermediate complexity. We first carried out two sets of simulations in which we increased the albedo of all land areas between ±20 • and ±45 • latitude respectively. The results of these simulations indicate a long-term global cooling effect of 3 × 10-15 K for each 1 m 2 of a surface with an albedo increase of 0.01. This temperature reduction corresponds to an equivalent CO 2 emission reduction of about 7 kg, based on recent estimates of the amount of global warming per unit CO 2 emission. In a series of additional simulations, we increased the albedo of urban locations only, on the basis of two independent estimates of the spatial extent of urban areas. In these simulations, global cooling ranged from 0.01 to 0.07 K, which corresponds to a CO 2 equivalent emission reduction of 25-150 billion tonnes of CO 2 .

Increasing surface reflectivity in urban areas can decrease ambient temperature, resulting in reducing photochemical reaction rates, reducing cooling energy demands and thus improving air quality and human health. The weather research and forecasting model with chemistry (WRF-Chem) is coupled with the multi-layer of the urban canopy model (ML-UCM) to investigate the effects of surface modification on urban climate in a two-way nested approach over North America focusing on Sacramento, Houston, and Chicago during the 2011 heat wave period. This approach decreases the uncertainties associated with scale separation and grid resolution and equip us with an integrated simulation setup to capture the full impacts of meteorological and photochemical reactions. WRF-ChemV3.6.1 simulated the diurnal variation of air temperature reasonably well, overpredicted wind speed and dew point temperature, underpredicted relative humidity, overpredicted ozone and nitrogen dioxide concentrations, and underpredicted fine particular matters (PM_2.5). The performance of PM_2.5 is a combination of overprediction of particulate sulfate and underprediction of particulate nitrate and organic carbon. Increasing the surface albedo of roofs, walls, and pavements from 0.2 to 0.65, 0.60, and 0.45, respectively, resulted in a decrease in air temperature by 2.3 °C in urban areas and 0.7 °C in suburban areas; a slight increase in wind speed; an increase in relative humidity (3%) and dew point temperature (0.3 °C); a decrease of PM_2.5 and O₃ concentrations by 2.7 µg/m³ and 6.3 ppb in urban areas and 1.4 µg/m³ and 2.5 ppb in suburban areas, respectively; minimal changes in PM_2.5 subspecies; and a decrease of nitrogen dioxide (1 ppb) in urban areas.