Revealing spatial and temporal patterns of residential cooling in Southern California through combined estimates of AC ownership and use

Air conditioning (AC) is an important tool for combatting the adverse health effects of heat, but its use can also drive surges of high demand for electricity. To better understand these effects, there is a need for non-intrusive methods of estimating AC access and operation. In this study, we use a novel methodology to identify residential AC ownership rates using smart meter data from 200,000 customers in Southern California, and find that 79 % of all customers in the region have AC. In contrast to previous methods, we classify AC ownership using hourly, rather than daily, electricity consumption records and directly account for the potential presence of electric heating. We then adapt and apply an algorithm to determine in which hours these households operate their AC. We estimate that the average customer runs their AC during 8.3 % of all hours in the two-year study period, but census-tract level averages range from 1 to 23 % of all hours. Lastly, we combine our estimates of AC ownership and use to analyze cooling behavior spatially and temporally, and are able to identify pockets of high cooling demand, areas lacking in access or the ability to use their AC, and potential targets for cooling-related DR programs.