Evaluation of MM5 and Eta-10 precipitation forecasts over the Pacific northwest during the cool season

Precipitation forecasts from the Pennsylvania State University-National Center for Atmospheric Research Mesoscale Model (MM5) and NCEP's 10-km resolution Eta Model (Eta-10) are verified over the Pacific Northwest in order to show the effects of increasing horizontal resolution, the spatial variations in model skill across the region, and the relative differences in performance between the two modeling systems. The MM5 is verified at 36- and 12-km resolution for 9 December 1996 through 30 April 1997 using approximately 150 cooperative observer and National Weather Service precipitation sites across the Pacific Northwest. A noticeable improvement in bias, equitable threat, and root-mean-square (rms) error scores occurs as the horizontal resolution is increased. The spatial distribution of bias and equitable threat scores across Washington and Oregon indicate that the 12-km MM5 generates too much precipitation along the steep windward slopes and not enough precipitation in the lee of major barriers. The MM5 results were compared with Eta-lO forecasts from 7 January 1997 through 30 April 1997. The Eta-10 overpredicts precipitation along the windward slopes even more than the 12-km MMS. As with the MM5, the Eta-lO forecasts excessive rainshadowing downwind of major barriers. Overall, the Eta-lO has lower rms errors than the 12-km MM5 at low precipitation thresholds, while the MM5 does significantly better than the Eta-lO for higher thresholds (>2.54 cm in 18 h).