The impact of white roof coatings on energy loads in three nonresidential buildings in Sacramento was assessed using a new approach to statistical analysis. Hourly metered loads were designed to isolate the effects of white roof coatings on building cooling loads from changes in cooling loads due to variations in outdoor temperatures. For this analysis, the basic multiple linear regression model used to weather normalize energy consumption data was expanded to include hourly solar radiation or insolation levels as explanatory variables, along with explanatory variables representing outdoor temperatures. With this statistical modeling approach, the effect of white roofs can be isolated from changes due to variations in outdoor temperatures based on the decrease in regression coefficients representing the effect of solar radiation levels on cooling loads after application of white roofs. Results indicate that the effect of solar insolation levels on cooling energy consumption was significantly decreased after the application of white roofs in all three buildings. Savings estimates based on this approach range from 17% to 39% of total cooling loads, or .35 kWh to .68 kWh per square foot of treated roof area per year.

Units: I-P