Modeling the Meteorological and Energy Impacts of Urban Heat Island Control in the U.S.

H. Taha, S. Gabersek, S.J. Konopacki

A mesoscale meteorological modeling study was undertaken to assess the impacts of urban heat island control strategies on urban meteorology and energy use in several U.S. cities. We modeled the impacts of large-scale use of high-albedo (reflective) materials in building and urban surfaces and the impacts of urban reforestation. Nine large regions, about 100 000 km2 each, were simulated. For each region, domain-averaged direct and indirect impacts of these strategies on energy use were quantified.

In the past, simulations of the direct (small-scale) effects of heat island control strategies on energy use have been performed on an annual basis using the DOE-2 building energy analysis program. The indirect (urban-scale) effects have been simulated and assessed over much shorter time-scales, e.g., a few days in summer, because mesoscale meteorological models cannot be easily adapted to run for an entire year. In this study, a methodology was developed to extrapolate "episodic" simulations of the indirect effect to include an entire year at each location. This allowed us to account for the direct and indirect impacts of urban heat island mitigation strategies on heating energy use in winter as well as cooling energy use in summer.

The Colorado State University Mesoscale Model (CSUMM) was used to simulate the regional meteorology and its modification. Episodic simulations of a few days at a time were performed for each region. To assess the direct and indirect energy implications of large-scale albedo and vegetation cover modifications, the DOE-2 program was used to simulate four prototypical buildings in each region using weather data generated by the meteorological model.

The mesoscale meteorological modeling results indicate that heat islands in Atlanta, Chicago, and New York may reach 2-3°C higher than the suburbs in the afternoon in summer, but only 1.5°C higher in Dallas and Phoenix. The Houston, Washington D.C., and Philadelphia heat islands are on the order of ~2°C, whereas Miami did not appear to have any heat islands, probably due to its coastal, open location. These heat islands were the direct consequences of low vegetation cover, surface moisture, and albedo and higher anthropogenic heating in urban areas. In terms of urban cooling through implementing high-albedo materials and urban forest strategies (combining impacts of reflective materials and trees), this mesoscale modeling study suggests that New York, Philadelphia, Washington D.C., and Atlanta have the highest potential for cooling (up to 2°C reduction in air temperature). Dallas, Chicago, Houston, and Phoenix follow, with temperature reduction of up to 1.5°C. Miami has the smallest cooling potential, up to 1°C reduction. The modeling results also suggest that cities with large surface areas available for modification are the ones that benefit most from heat island mitigation strategies, e.g., New York (see Figure).

Building simulations with the DOE-2 program indicate that energy savings on the order of 3-35% are attainable with modifications in albedo and vegetation cover. The savings at each location depend on building type and thermal integrity, weather, and other location-specific conditions. Simulations indicate that the percentage of total energy savings attributable to the indirect effects is between 10% and 40% (i.e., the direct effects contribute 60% to 90% to total savings). However, the indirect effects in these simulations are probably underestimated because the meteorological model simulates conditions in the boundary layer (e.g., above roof level) not in the canopy layer (e.g., under a vegetation stand or within street canyons) where air temperature is more responsive to changes in surface properties. Future work on these models should allow us to obtain "below-the-canopy" estimates of reductions in temperature and, thus, savings in energy use. An important conclusion we reached based on the results of the modeling work is that wintertime penalties in heating energy use are very small or negligible in most cases.

Reference

Gabersek S, Taha H. A Preliminary Multi-City Assessment of the Impacts of Increased Urban Albedo and Vegetation on Regional Meteorology and Energy Use. Lawrence Berkeley National Laboratory Report No. LBL-37887, 1995.

Figure. Simulated cooling (°C) at 3 p.m. on a typical July 25 in the New York modeling region resulting from large-scale use of reflective materials and urban trees. Blue areas indicate a temperature decrease; the darker the blue, the greater the temperature decrease. A large area has cooled by as much as 2°C over New York City, Long Island, Newark, and Philadelphia. (Red areas indicate a slight increase in temperature.)

| Return to Contents | Return to Previous Paper | Next Paper |