Climatological and radar-indicated characteristics of United States extreme rain events
Date
2003
Authors
Schumacher, Russ, author
Journal Title
Journal ISSN
Volume Title
Abstract
This study examines the radar-indicated structures and other features of extreme rain events in the United States over a three-year period. A rainfall event is defined as "extreme" when the 24-h precipitation total at one or more stations surpasses a threshold that varies spatially based on the frequency of heavy rain in that location. Using the National Weather Service high-resolution rain gauge network, and eliminating bad precipitation data, this definition yields 193 such events from 1999 to 2001 in the area east of the Rocky Mountains, excl ding Florida. The monthly frequency distribution of extreme rain events generally agrees with previous studies of precipitation systems. There is an overall maximum in June, July, and August, though events occur in every month of the year. In the northern part of the country, extreme rainfall is almost exclusively confined to the warm season. In contrast, the distribution in the south indicates that extreme rainfall there is less dependent on season. Two-km national composite radar reflectivity data are used to examine the structure and evolution of each extreme rain event. In the northern and Great Plains states, almost all of the extreme rain events are associated with mesoscale convective systems (MCSs), while in the northeast and southeast, synoptic-scale weather systems and tropical cyclones play a much greater role. In total, 59% of the total number of events are associated with MCSs. The heavy rain from extreme-rain-producing MCSs typically begins in the evening, peaks sometime after dark, and dissipates or moves away in the early morning hours. While a wide variety of organizational structures (as indicated by the reflectivity data) were seen among the MCS cases, two patterns of organization were observed most frequently. The first pattern has a line with "training" convective elements and an adjoining region of stratiform rain. The second has a back building convective line/cluster with a parallel region of stratiform rain downstream. The organization of each type of MCS is conducive to large rainfall accumulations. Composite analysis of RUC- 2 model analyses reveals that training line/adjoining stratiform (TL/AS) systems typically form under deep atmospheric moisture on the cool side of a pre-existing slow-moving surface boundary, while back building with parallel stratiform (BPS) MCSs also occur in a very moist environment but are more dependent on mesoscale and storm-scale processes than on pre-existing synoptic boundaries. Two case studies are also presented that further reinforce the conclusions drawn from the composite analysis.
Description
Fall 2003.
Rights Access
Subject
Severe storms -- United States
Doppler radar
Flood forecasting