The goal of the Photosynthetically available radiation (PAR) for Primary Production (3P) project is to provide robust, complete, and user-friendly satellite radiation products for ecosystem modeling, carbon cycle investigations, and climate change monitoring. A specific objective is to design and distribute a daily PAR product from MERIS and potentially the recent OLCI. In view of this, a PAR algorithm, based on the NASA Ocean Biology Processing Group (OBPG) operational algorithm, has been developed. The algorithm takes into account statistical diurnal variability of clouds using 3-hourly International Satellite Cloud Climatology (ISCCP) data. The PAR modeling, simplified to accommodate the information available, is evaluated using a Monte Carlo tool that simulates the satellite radiance and corresponding daily PAR. The daily PAR estimates obtained from reduced resolution (i.e., 1 km) MERIS data are evaluated against in situ measurements routinely collected from fixed buoys and platforms, namely BOUSSOLE in the Mediterranean Sea, CCE- 1 and -2 off the West coast of the United States, and COVE in the coastal Atlantic Ocean. The agreement between estimated and measured values is good on a daily time scale and substantially improved on a monthly time scale, with a bias of 2.7 (7.7%) E/m2/day and RMS errors of 8.5 (24.9%) and 4.5 (12.9%) E/m2/day. The bias is reduced significantly (by 1.8%) when using diurnal cloud climatology. Overestimation in cloudy conditions is partly explained by decoupling the clear atmosphere from the cloud/surface layer. Large gaps in regions affected by sun glint (not processed because incorrectly interpreted as cloudy) are adequately filled in the monthly PAR imagery. The statistical performance is satisfactory for long-term studies of aquatic primary production, especially in view of the much larger uncertainties on the fraction of PAR absorbed by live algae and the quantum yield of carbon fixation.