The future market for residential photovoltaic systems: New perspectives on the rooftop resource base

The resource base of one-family houses with roofs suitable for photovoltaic applications can be estimated using a new and efficient sampling strategy combined with a statistical model. Levels of residential housing density are good predictors of rooftop suitability and can be used to make out-of-sample forecasts for a variety of geographic areas. The model is particularly useful for making forecasts for census tracts within large metropolitan areas. A large-scale field survey of rooftops in southeastern Pennsylvania indicates that a much higher proportion of roofs are suitable for rooftop systems than previously thought (in the Northeast). The survey suggests important differences from previous assumptions for such roof characteristics as area, slope, aspect, and the frequency of roof obstructions. Overall, the rooftop survey provides a new paradigm for understanding the rooftop resource base; this paradigm rests on the dramatic difference between the central city and the suburbs in the proportion of houses with suitable roofs. The survey also demonstrates the critical role played by landscaping practices within a single urban area