The ECMWF 40-year reanalysis (ERA40) is proceeding in three streams, and the most recent period, 1989-2000, will be complete this year. Surface energy and water budgets, averaged over river basins, are computed and archived during the analysis cycle. We use these and precipitation observations to assess the systematic biases in the surface energy and water budget of both the ECMWF reanalyis and the NASA-DAO finite volume atmospheric GCM for five Mississippi sub-basins. The fvGCM was run with 1x1.25 deg. horizontal resolution for the 15 years, 1986-2000, using observed varying sea surface temperatures. The fvGCM atmospheric physics are taken from NCAR CCM3, using the NCAR land-surface model. The fvGCM has much higher net shortwave in summer than ERA-40, and this gives generally higher evaporation in summer. ERA-40 has higher evaporation in winter, probably because it has no seasonal vegetation cycle. The fvGCM has a large cold surface temperature bias in winter in comparison to ERA-40. There are big differences in the partition between large-scale (LSP) and convective precipitation (CP) between the two models. LSP, which dominates the cool season, is much smaller in the fvGCM, and its summer CP is higher than in ERA-40. The spinup of LSP in ERA-40 is quite large. For some basins such as the Upper Mississippi and the Ohio-Tennessee, the 12-24 hr precipitation in ERA-40 exceeds that observed by 20-30% in some seasons.