The atmospheric moisture budget is evaluated for the region 70°N to the North Pole using reanalysis data sets of the European Centre for Medium-Range Weather Forecasts (ECMWF; ERA: ECMWF Re-Analysis) and the collaborative effort of the National Centers for Environmental Prediction (NCEP) and the National Center for Atmospheric Research (NCAR). For the forecast fields of the reanalyses, the ERA annually-averaged P-E (precipitation minus evaporation/sublimation) field reproduces the major features of the basin perimeter as they are known, while the NCEP/NCAR Reanalysis forecast fields contain a spurious wave pattern in both P and E. Comparisons between gauge data from Soviet drift camp stations and forecast P values of the reanalyses show reasonable agreement given the difficulties (i.e., gauge accuracy, translating location). When averaged for 70°N to 90°N, the ERA and NCEP/NCAR forecast P-E are similar in the annual cycle and interannual variability. Average reanalysis forecast values of E for the north polar cap are found to be 40% or more too large based on comparisons using surface latent heat flux climatologies.
Reasonable agreement is found at point locations between the derived atmospheric moisture fluxes and in situ data from the Historical Arctic Rawinsonde Archive (HARA). Differences with a synthesized average moisture flux across 70°N from rawinsonde data and the reanalysis data occur in the presence of rawinsonde network problems. It is concluded that deficiencies exist in the rawinsonde depiction of the summertime meridional moisture transport, both in the long-term mean and the interannual variability. However, it remains to be seen whether the rawinsonde estimate can be rectified with a different method. For 70°N-90°N, the time series of annual moisture convergence (P-E) values from the ERA and NCEP/NCAR are very similar. There is significant disagreement with values obtained from the rawinsonde network. Annual values obtained from forecast fields are much lower for both reanalyses, by about 60%, indicating severe non-closure. The interannual variability is very similar for both methods and both reanalysis data sets, however. In comparison with other studies, reanalyses moisture convergence values are more reasonable. A synthesis of the reanalysis moisture budget values and previous studies suggests a value of 18.9±2.3 cm yr-1 for the north polar cap.