Impact of climate change on U.S. building energy demand: sensitivity to spatiotemporal scales, balance point temperature, and population distribution.

Past assessments of climate change impacts on building energy consumption have typically neglected spatial variations in the 'balance point' temperature, population distribution effects, and the extremes at smaller spatiotemporal scales where the impacts of climate change are most pronounced. Here we test the impact of these limitations through a sensitivity analysis in the Contiguous United States. Though national/annual total source energy consumption differences between the 2080-99 time period and the present are less than 2 %, we find changes at the state/month scale that are much larger with summer electricity demand increases exceeding 50 % and spring non-electric energy declines of 48 % by the end of the century. The use of a fixed 18.3 °C (65 °F) balance point temperature, versus a more representative state-specific value, leads to an overestimate of the energy consumption changes in most states with a maximum change in the state of Oregon of almost 14 percentage points. Finally, projected population redistribution, when combined with the spatial pattern of climate change, exacerbates the building energy consumption impacts, further increasing source energy consumption in some states (max = +5.3 percentage points) and further diminishing energy consumption declines in others (max = −8.2 percentage points). When integrated over the U.S., the intersection of projected population distribution changes and climate change shifts future building energy consumption from a net decrease to a net increase. [ABSTRACT FROM AUTHOR]