City-scale energetics: window on adaptive thermal insulation in North American cities

Previous research demonstrated that cities are similar to individual mammals in their relationship between the rate of energy use for heating and outdoor air temperature (T(a)). At T(a)s requiring heating of indoor living spaces, the energy-T(a) plot of a city contains information on city-wide thermal insulation (I), making it possible to quantify city-wide I by use of the city as the unit of measure. We develop methods for extracting this insulation information, deriving the methods from prior research on mammals. Using these methods, we address the question: in North America, are cities built in particularly cold locations constructed in ways that provide greater thermal insulation than ones built in thermally more moderate locations? Using data for 42 small and medium-size cities and two information-extraction methods, we find that there is a statistically significant inverse relationship between city-wide I and T(10-year), the average city T(a) over a recent 10-year period (range of T(10-year): − 11 to 26 °C). This relationship represents an energy-conserving trend, indicating that cities in cold climates have greater built-in thermal insulation than cities in warm climates. However, the augmentation of insulation in cold climates is only about half as great as would be required to offset fully the increased energy cost of low T(a)s in a cold climate, and T(10-year) explains just 5–11% of the variance in measured insulation, suggesting that cities in North America vary greatly in the extent to which thermal insulation has been a priority in city development. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00360-021-01411-8.