In this paper a clear-sky shortwave closure analysis is presented for the Baseline Surface Radiation Network (BSRN) site of Cabauw, Netherlands (51.97°N, 4.93°E). The analysis is based on an exceptional period of fine weather during the first half of May 2008, resulting in a selection of 72 comparisons, on 6 days, between BSRN measurements and Doubling Adding KNMI (DAK) model simulations of direct, diffuse, and global irradiances. The data span a wide range of aerosol properties, water vapor columns, and solar zenith angles. The model input consisted of operational Aerosol Robotic Network (AERONET) aerosol products and radiosonde data. The wavelength dependence of the aerosol optical thickness, single scattering albedo, and asymmetry parameter was taken into account. On the basis of these data, excellent closure was obtained: the mean differences between model and measurements are 2 W/m2 (+0.2%) for the direct irradiance, 1 W/m2 (+0.8%) for the diffuse irradiance, and 2 W/m2 (+0.3%) for the global irradiance. The good results were obtained because of proper specification of the DAK model input and the high quality of the AERONET and BSRN measurements. The sensitivity of the achieved closure to uncertainties in the aerosol optical thickness, single scattering albedo, and asymmetry parameter was examined. Furthermore, several sensitivity experiments related to the wavelength dependence of the aerosol optical properties and the treatment of water vapor were performed. It appeared that a correct description of the wavelength dependence of the aerosol optical properties is important for achieving broadband closure. However, broadband closure can also be obtained by means of using spectrally averaged values of the single scattering albedo and the asymmetry parameter. Cancellation of errors in different parts of the solar spectrum also contributes to the achieved closure