Comparison of techniques for in situ nondamaging measurement of solar reflectances of low-slope roof membranes

With the implementation of the Energy Star Roof Products Program by the U.S. Environmental Protection Agency and the U.S. Department of Energy, techniques are especially needed that yield in situ measurements of the average solar reflectance of roof surfaces without damage to them. This paper presents results of limited field surveys with two types of instruments that permit such measurements. Solar reflectances on a scale from 0 to 1 were obtained by the established laboratory technique for five samples covering the range exhibited by low-slope roofs and coating systems for them. Based on these results, the average bias for one instrument, a portable solar spectrum reflectometer using a built-in light source, was +0.003. The maximum bias for the five samples was ±0.02. Scatter of readings over a roof area with this instrument depends upon characteristics of the specific surface. Scatter can be as little as ±0.001 but is typically more than ±0.02. The other instrument uses a pyranometer and is operated by recording the responses when the pyranometer faces the sun and when it is inverted facing the surface of interest. The reflectance is the ratio of the response when inverted to the response facing the sun. For a variety of roof surfaces, the average of readings with both instruments agreed within 95% confidence intervals of ±0.02 to ±0.06, calculated as ±t·s.d., where t is the t-statistic for the number of measurements and s.d. is the measurement standard deviation.