A companion article explored how solar reflectance varies with surface orientation and solar position, and found that clear sky air mass 1 global horizontal (AM1GH) solar reflectance is a preferred quantity for estimating solar heat gain. In this study we show that AM1GH solar reflectance R(g,0) can be accurately measured with a pyranometer, a solar spectrophotometer, or an updated edition of the Solar Spectrum Reflectometer (version 6). Of primary concern are errors that result from variations in the spectral and angular distributions of incident sunlight. Neglecting shadow, background and instrument errors, the conventional pyranometer technique can measure R(g,0) to within 0.01 for surface slopes up to 5:12 [23 degrees], and to within 0.02 for surface slopes up to 12:12 [45 degrees]. An alternative pyranometer method minimizes shadow errors and can be used to measure R(g,0) of a surface as small as 1 m in diameter. The accuracy with which it can measure R(g,0) is otherwise comparable to that of the conventional pyranometer technique. A solar spectrophotometer can be used to determine R(g,0)*, a solar reflectance computed by averaging solar spectral reflectance weighted with AM1GH solar spectral irradiance. Neglecting instrument errors, R(g,0)* matches R(g,0) to within 0.006. The air mass 1.5 solar reflectance measured with version 5 of the Solar Spectrum Reflectometer can differ from R(g,0)* by as much as 0.08, but the AM1GH output of version 6 of this instrument matches R(g,0)* to within about 0.01. g. (C) 2010 Elsevier Ltd. All rights reserved.