In this work, we describe a system to estimate virtually all the components of the land-surface energy budget from space-based data sources and complementary in-situ synoptic data (rawinsondes), coupled with a diagnostic model of the land surface and a mesoscale numerical forecast model. The satellite inputs to the system are incident solar and net longwave radiation streams and time-changes of radiometric surface temperature (all derived from geostationary [GOES] satellite data), and vegetation indices from AVHRR data. These data are inputs to a diagnostic Atmosphere-Land Exchange Model Inverse (ALEXI), which also uses atmospheric profile information obtained from the analysis component of the CIMSS Regional Assimilation System (CRAS). The prognostic component of the CRAS provides Atmospheric Boundary Layer (ABL) wind speeds and other information to the ALEXI system at several observation times. With these data, ALEXI diagnoses the land surface latent and sensible heat fluxes, as well as soil conduction. Because land-surface radiometric temperatures are required by the ALEXI method, it is only possible to diagnose the latent, sensible and soil fluxes in clear observing periods. However, during these clear periods, surface moisture indices are derived and subsequently adjusted during cloudy periods with evaporation amount based on the net radiation (again from satellite data) and atmospheric conditions from the forecast model, so that a continuous record of the surface energy budget is maintained. In this paper, we describe this system (now running in real time over the continental US at a resolution of 10 km), as well as giving examples of results from the method over different space and time scales.