Radiative transfer calculations of UV irradiance from total ozone mapping spectrometer (TOMS) satellite data are frequently overestimated compared to ground-based measurements because of the presence of undetected absorbing aerosols in the planetary boundary layer. To reduce these uncertainties, an aerosol UV absorption closure experiment has been conducted at the National Aeronautics and Space Administration/Goddard Space Flight Center (NASA/GSFC) site in Greenbelt, Maryland, using 17 months of data from a shadowband radiometer [UV-multifilter rotating shadowband radiometer (UV-MFRSR), U.S. Department of Agriculture (USDA) UV-B Monitoring and Research Network] colocated with a group of three sun-sky CIMEL radiometers [rotating reference instruments of the NASA Aerosol Robotic Network (AERONET)]. We describe an improved UV-MFRSR on-site calibration method augmented by AERONET-CIMEL measurements of aerosol extinction optical thickness (τa) interpolated or extrapolated to the UV-MFRSR wavelengths and measurement intervals. The estimated τa is used as input to a UV-MFRSR spectral-band model, along with independent column ozone and surface pressure measurements, to estimate zero air mass voltages V0 in three longer wavelength UV-MFRSR channels (325, 332, 368 nm). Daily mean 〈V0〉, estimates and standard deviations are obtained for cloud-free conditions and compared with the on-site UV-MFRSR Langley plot calibration method. By repeating the calibrations on clear days, relatively good stability (±2% in 〈V0〉) is found in summer, with larger relative changes in fall-winter seasons.