Fine-mesh models, such as the eta model, are producing increasingly detailed predictions about mesoscale atmospheric motions. Mesoscale systems typically produce stronger vertical motions than do synoptic-scale storms, making it more difficult for forecasters to assess the strength of the latter's dynamics when the signals are overwhelmed by mesoscale processes. This paper describes a method for extracting synoptic-scale information from mesoscale model data. Predicted height fields from the 29-km eta model are investigated to determine the filtering and smoothing requirements necessary to resolve synoptic-scale patterns of vertical motions using quasigeostrophic (QG) diagnostics. The selected late-fall case includes a jet stream that enters the continent over the Pacific Northwest, resulting in orographically induced troughs in the lee of the Cascade Range and Rocky Mountains. Gravity waves are found to emanate from this region in arcs that reach Hudson Bay to the northeast and extend to the Caribbean in the southeast. Individual gravity wave crests (~240 km apart) are of sufficient amplitude (5 to 10 m at 500 mb) to dominate the expected synoptic-scale vertical motions by two orders of magnitude. A numerical filter based on a two-dimensional diffraction function is designed, tested, and found to eliminate the influence of the gravity waves effectively. The filtered model data are then able to reveal synoptic-scale vertical motion patterns in all areas except the vicinity of the lee troughs, which still dominate QG forcing near the jet axis.
*National Oceanic and Atmospheric Administration, Forecast Systems Laboratory, Boulder, Colorado.
+Cooperative Institute for Research in the Atmosphere, Fort Collins, Colorado.