Remote sensing products, such as the fraction of reflected solar radiation flux, as well as the amount of radiation absorbed in the photosynthetically active spectral region and the Leaf Area Index (LAI), are operationally available from Space Agencies. Climate models may benefit from these products provided their one dimensional (1-D) radiation transfer schemes effectively represent the three dimensional (3-D) effects implied by the internal spatial variability of vegetation canopies, e.g., the leaf area density, at all scales and resolutions involved (say from 1 to 100 kilometers). Failing to do so leads to inherent inconsistencies between the domain-averaged reflected and absorbed fluxes, and the implied Leaf Area Index. We propose a comprehensive approach which introduces a parameterization of the internal variability of the LAI in the 1-D representation of the radiation scheme, called a domain-averaged structure factor, and provides a description of the radiant fluxes fully consistent with the LAI specified by remote sensing. We take this opportunity to revisit and update the two-stream formulations implemented in climate models to accurately estimate the fractions of radiation absorbed separately by the vegetation canopy and the underlying surface. This is achieved by isolating the contributions of the vegetation canopy alone, the background as seen through the canopy gaps and the multiple scattering between the vegetation layer and the background. The performance of this formulation is evaluated against results from Monte Carlo simulations relative to explicit realistic 3-D canopies to show that the proposed scheme correctly simulates both the amplitude and the angular variations of all radiant fluxes with respect to the solar zenith angle.