It is examined the seasonal variability of stationary moisture fluxes in the NASA/NSIPP atmospheric model. Ensemble number five of an AMIP run of the model is used while NCEP/NCAR and ECMWF reanalyses data sets are used as targets for the simulation.

Although the model depicts the general characteristics of monthly and seasonal vertically integrated moisture fluxes, mostly to the north of 30N, it fails to simulate main features of the contrasting seasons of summer and winter appearing in the reanalyses. The model does not have both a well defined low level jet in summer transporting moisture from the Gulf of Mexico into the Great Plains region nor a well established moisture flux from the midlatitude eastern Pacific into the northwest portion of the U.S. during winter. In consequence the mean moisture fluxes from the Pacific and the Gulf of Mexico are weak.

It is suggested that those inaccuracies by the model are due to misrepresentations of the surface winds related to the semi permanent midlatitude anticyclones. In particular, the existence of known bias in precipitation in the eastern tropical Pacific is the ultimate cause of weak southerly moisture fluxes from the Gulf of Mexico into the United States.