While global observational estimates of energy fluxes in and out of the climate system at the top of atmosphere (TOA) are now available with considerable accuracy from recent satellite programs, similar reference values are more difficult to establish for the surface energy fluxes, which cannot be directly measured from space. This is reflected in greatly diverging global estimates of the surface energy balance components that have been published over the years, or simulated in global climate models. Since the mid-1990s, accurate direct measurements become increasingly available from the networks of surface radiation stations, which allow to better constrain the energy fluxes at the Earth's surface. In parallel, satellite-derived products of surface fluxes profit from the great advancement in space-born observation systems that became operational since the turn of the millennium. As a consequence, recent independent global estimates of the surface radiation components based on surface and satellite data sources have converged to within a few watts per square meter. This suggests that we are approaching a stage where we are not only confident in the magnitudes of the global energy balance components at the TOA but also increasingly so at the surface. These recent estimates may also be able to reconcile formerly disputed inconsistencies between global energy and water cycle estimates. Remaining challenges include the accurate determination of representative surface albedos and skin temperatures in the calculation of surface shortwave absorption and upward longwave emission, respectively, as well as the partitioning of surface net radiation into the nonradiative fluxes of sensible and latent heat.