Extended Abstract (1.6M)2A.5
Initial analysis of model guidance for the Colorado Front Range barrage of snowstorms Nov. 2006-Feb. 2007
Douglas A. Wesley, UCAR/COMET, Boulder, CO; and R. A. Pielke, S. D. Jascourt, W. Bua, D. Barjenbruch, E. Szoke, and G. Poulos
During late Nov. 2006 through late Feb. 2007 (press time) the Colorado Front Range was subjected to a barrage of snow events, some high-impact in nature, and some that were not only climatologically anomalous but difficult to forecast. Snowfall was significant in the region for at least 9 events during the period, and aggregate totals for many locations on the west side of the urban corridor easily exceeded 4 meters for a period that is prior to the climatologically snowiest season (spring). The heaviest snowfall of Dec. 20-21 shut down major ground and air transportation systems throughout the region for 2+ days. Other high plains regions also experienced devastating winter storm conditions in this storm.
The transition of the NOAA/NWS primary numerical guidance from the Eta model to the WRF-NMM model resulted in this period being an adjustment phase for operational forecasters, as the lower-level terrain-following grids in the NAM now utilize the sigma-coordinate rather than the Eta-coordinate. NAM12 and GFS40 forecasts were the primary operational tools on the AWIPS systems. Problems with the new NAM resulted in NCEP implementing changes on Dec. 19, 2006, perhaps affecting the forecasts of the event occurring immediately after this date. Experimental 5km NMM guidance was available via the Web, but these model results are not the emphasis of this study.
The authors noted during the forecast periods for several of these storms that performance of the NAM12 was inconsistent, and occasionally poorer than that of the relatively poorly-resolved GFS40 spectral model. This investigation takes an initial look at 36-72 hour guidance and verification for these storms, both on a case-by-case basis and in aggregate. Model pressure-level data (primarily height and winds) as well as some moisture fields are evaluated during the 2-3 day period ahead of the storms. Examples of actual forecasts (data and watches/warnings) will be shown, as well as insight towards what model guidance the forecasters were emphasizing and why. Ensemble data (SREF) available from NCEP for these models will be evaluated, as presentation time allows.
Generally, when a potentially significant winter storm is lurking, 2-3 day forecasts are critical to public, emergency, and private sector officials who require some accuracy in order to perform their jobs. Recently, debate in the NWS and the modeling research community has centered on whether the human forecaster adds value during the 2-4 day forecast period. Numerical guidance improvements have generally enabled the community to make strides towards automation. This study will attempt to provide some relevant baseline data for forecasting wintertime storms in complex terrain 2-3 days ahead of time.
Session 2A, Winter Weather
Tuesday, 26 June 2007, 10:30 AM-12:00 PM, Summit A
Previous paper Next paper