The Eta Model Surface Hydrologic Cycle of the Columbia and Colorado Basins

The surface hydrology of the United States western basins is investigated using the National Centers for Environmental Prediction Eta model operational forecasts. During recent years the model has been subject to changes and upgrades that affected positively its performance. We discuss these effects on the surface hydrologic cycle by analyzing the period June 1995-May 2003. Results are compared with available precipitation observations and land surface hydrologic estimates resulting from the Variable Infiltration Capacity (VIC) macroscale hydrologic model. A fairly large disparity between the observed precipitation estimates with and without orography correction is found in the long term area-averages over the Columbia basin and over the Colorado basin. The basin-averaged model precipitation correlates well with the observations at monthly timescales, but a 34% positive difference with respect to the not-orographically corrected precipitation is found over the Columbia basin; the difference is reduced to 2% when using the orographically corrected precipitation. The excess in the model forecast precipitation is mainly found during the winter months, and seems related both to the model's large scale precipitation component, but also to the underestimation of gauge-based precipitation measurements. The model bias over the Colorado basin is largest during summer, and it is suggested that the limited performance of the convective precipitation component is the one that leads to the underestimation of the precipitation. The mean fields of the hydrological variables in the Eta model are in qualitative agreement with those from the VIC model at regional-to-large scales. As expected, the largest differences are found near mountains and the western coastline. While the mean fields of precipitation, evaporation, runoff and normalized soil moisture are in general agreement, important differences arise in their mean annual cycle over the two basins: snow melt in the Eta model precedes that of VIC by two months, and this phase shift is also reflected in the other variables. Differences are largest during spring, and as the seasons progress toward late fall and into winter the estimates become closer. Promising improvements have been found in the model performance in terms of quality of the forecast precipitation and in the reduction of the residual term of the surface water balance. Both effects are most evident in the last three to four years, suggesting that at least similar (or better) quality will be found in studies based on the recently released North American Regional Reanalysis dataset.