BOULDER--Until now scientists have found it hard to predict which summer storms forming over the Rocky Mountains would produce giant, flood-prone storm systems in the Great Plains to the east. Now Andrew Crook (National Center for Atmospheric Research, or NCAR) and Donna Tucker (University of Kansas) may have found the key: the strength of intense downdrafts that emerge from the mountain storms and stir up severe weather downstream. Computer modeling to track these downdrafts and the cloud-level ice crystals that help produce them may eventually give forecasters the edge in predicting severe storm systems, and possibly flooding, over the plains. Crook and Tucker (the lead author) are publishing their results in the June issue of Monthly Weather Review. NCAR's primary sponsor is the National Science Foundation.
Most summertime floods across the Great Plains are caused by mesoscale convective systems (MCSs). These giant complexes often emerge from showers and thunderstorms that form over the Rocky Mountains. Tucker and Crook used the Pennsylvania State University/NCAR mesoscale model to simulate convection (showers and thunderstorms) and to test how different modes of mountain convection affect the likelihood of MCS formation downstream. In the model, they found that an MCS was most likely to form when a mass of rain-cooled air descended from the mountains, colliding with moist air on the plains and forcing it upward.
Although forecasters have seen this process unfold many times, it is still unclear whether a given day's mountain storms will be the right kind to trigger an MCS. Sometimes the initial storms lead to an MCS that can travel as far as Illinois; other times, the storms dissipate shortly after they move off the mountains. Tucker and Crook's modeling suggests that the strength of the rain-cooled outflow from the mountain storms is critical to downstream MCS development. Several factors play into the outflow strength, including the fall speed of ice crystals within the mountain storms.
Fine-scale modeling for better prediction
Even today's most sophisticated forecast models cannot peg mountain convection well enough to assess how it might trigger storm complexes downstream. However, under a new NSF grant, Tucker and Crook are using a finer-scale model built by NCAR scientist Terry Clark to look more closely at mountain convection and how it relates to the larger-scale atmospheric flow. Since the large-scale flow is routinely forecast by computer models, this new work could allow forecasters to better pinpoint a given day's mountain convection and where it might trigger large storm complexes on the plains. Tucker and Crook's work is supported by the University of Kansas and NSF.
One downpour leads to another: NCAR team pinpoints culprit
A typical MCS peaks in strength during the overnight hours and dissipates the next day. However, it may be followed by a second MCS the following night. Sometimes a slow-moving sequence of MCSs will extend over several days, causing torrential rains over a large area. If such a multiday sequence could be forecast, valuable lead time might be gained on flooding threats.
NCAR scientists Christopher Davis, Stanley Trier, and colleagues have gained new insight on a type of low-pressure center that connects one MCS to the next. This low is called a mesoscale convective vortex (MCV). With a core only 30 to 60 miles wide and 1 to 3 miles deep, an MCV is often overlooked in standard weather analyses. But Davis and Trier have found that MCVs play a key role in helping storms regenerate over two or more days.
Looking closely at satellite, upper-air, and radar observations from 1998, Davis and Trier found evidence of 17 separate MCVs over the central and eastern United States. Previous studies had found only two or three MCVs per year. The vortices appear most likely to persist when lower- and upper-level winds are relatively light. This allows the circulation to maintain its integrity for up to 12 hours after the storms dissipate. If other conditions are favorable, a new round of storms may cluster around the vortex. For example, one MCV triggered heavy rains in Texas on May 27, 1998; flooding in Arkansas early on the 28th; and additional flooding the following night in Mississippi. An MCV that moves into tropical waters, such as the Gulf of Mexico, can serve as the nucleus for a tropical storm or hurricane.
Currently, it's difficult to spot and track mesoscale convective vortices from upper wind observations alone, due to their small size. However, a technique developed by NCAR's John Tuttle calculates winds using cloud movements observed by satellite in order to spot MCVs and other features. This promising technique, along with better observations and models, could make it practical for forecasters to use MCVs as a guide to predict locations of heavy rain. Davis and Trier's work is supported by NASA and the U.S. Weather Research Program, which is examining forecast tools for heavy precipitation.
NCAR is managed by the University Corporation for Atmospheric Research, a consortium of more than 60 universities offering Ph.D.s in atmospheric and related sciences.
Writer: Bob Henson
Note to Editors:
Visuals: Images are available at ftp://ftp.ucar.edu/communications.
Filename(s): mcsradar.tif, downdraft.tif, sat.tif, satwind.tif. Captions
are at the Web address below.
UCAR and NCAR news: http://www.ucar.edu/publications/newsreleases/1999.
To subscribe via e-mail send name, title, affiliation, postal address, fax, and phone number to butterwo@ucar.edu.