In this study, the vertical fluxes of the global water cycle between the surface and the atmosphere - precipitation and evaporation - are explored using monthly estimates from the Global Precipitation Climatology Project (GPCP), retrieval estimates of ocean evaporation from Special Scanning Microwave Imager (SSM/I) data, and model estimates of continental evaporation from retrospective runs under the Global Land Data Assimilation (GLDAS) framework and similar global land-model simulations. The analysis will assess the extent to which these global observations/estimates of precipitation and evaporation reasonably balance in the global water-cycle budget (under the supposition that trends/variations in the global storages of vapor/liquid/frozen water are of considerably smaller magnitude), characterize the relative contributions of the ocean and land on the global precipitation and evaporation rates, and characterize the mean and interannual variability of the global precipitation and evaporation fluxes (i.e. the "rate" of the global water cycle). In addition, correlation diagnostics are performed to assess the extent to which variations in the "rate" of the global water cycle can be associated with large-scale climate oscillations (e.g. AO, PNA, and ENSO) and global changes.

In an effort to determine whether general circulation models (GCMs) can reproduce the above empirically-based characterizations of the global water cycle, a suite of ensemble simulations from NASA's Seasonal to Interannual Prediction Project (NSIPP) will be analyzed accordingly. These long-term simulations were run with prescribed interannually varying sea-surface temperatures and span the period 1930-2000. As such, these runs overlap the records of the aforementioned observations/estimates of global precipitation and evaporation and will therefore be judged against the empirical diagnoses. However, a complementary set of statistics will also be presented that covers the remaining decades of the simulation.