Synoptic Control of Mesoscale Precipitating Systems in the Pacific Northwest

This research examines whether an adequate representation of flow features on the synoptic scale allows for the skillful inference of mesoscale precipitating systems. The focus is on the specific problem of landfalling systems on the west coast of the United States for a variety of synoptic types that lead to significant rainfall. The methodology emphasizes rigorous hypothesis testing within a controlled hindcast setting to quantify the significance of the results. The role of lateral boundary conditions is explicitly accounted for by the study. The hypotheses that (a) uncertainty in the large-scale analysis and (b) upstream buffer size have no impact on the skill of precipitation simulations are each rejected at a high level of confidence, with the results showing that mean precipitation skill is higher where low analysis uncertainty exists and for small nested grids. This indicates that an important connection exists between the quality of the synoptic information and predictability at the mesoscale in this environment, despite the absence of such information in the initialization or boundary conditions. Further, the flow-through of synoptic information strongly constrains the evolution of the mesoscale such that a small upstream buffer produces superior results consistent with the higher quality of the information crossing the boundary. Some preliminary evidence that synoptic type has an influence on precipitation skill is also found. The implications of these results for data assimilation, forecasting, and climate modeling are discussed.