Lawrence Berkeley National Laboratory

The ability of a climate model to accurately simulate the urban cooling effect of raising street albedo may be hampered by unrealistic representations of street geometry in the urban canyon. Even if the climate model is coupled to an urban canyon model (UCM), it is hard to define detailed urban geometries in UCMs. In this study, we relate simulated surface air temperature change to canyon albedo change. Using this relationship, we calculate scaling factors to adjust previously obtained surface air temperature changes that were simulated using generic canyon geometries. The adjusted temperature changes are obtained using a proposed multi-reflection urban canyon albedo model (UCAM), avoiding the need to rerun computationally expensive climate models. The adjusted temperature changes represent those that would be obtained from simulating with city-specific (local) geometries. Local urban geometries are estimated from details of the city's building stock and the city's street design guidelines. As a case study, we calculated average citywide seasonal scaling factors for realistic canyon geometries in Sacramento, California based on street design guidelines and building stock. The average scaling factors are multipliers used to adjust air temperature changes previously simulated by a Weather Research and Forecasting model coupled to an urban canyon model in which streets extended from wall to wall (omitting setbacks, such as sidewalks and yards). Sacramento's scaling factors ranged from 2.70 (summer) to 3.89 (winter), demonstrating the need to consider the actual urban geometry in urban climate studies.