Monitoring the Storm's Path
The infrared satellite and radar reflectivity image (below) provide an excellent overview of the supercell thunderstorm over central Illinois around 8 PM CST on the 12th. By toggling between the two images, it is possible to see how the cloud tops on the infrared image (shown in white) correlate to the regions of high reflectivity values (shown in shades of yellow and red) on the radar image. The large, intense supercell thunderstorm over central Illinois is clearly evident on the both the infrared satellite and radar image.
The defining element of a supercell thunderstorm is its mesocyclone. A mesocyclone is a column of rotating air, usually from two to six miles in diameter, that is typically found in the right rear portion of the storm. It is the mesocyclone that is responsible for the formation of tornadoes. The Radar Summary from 0215Z on March 13th (815 PM CST on 3/12) displays those storms where radar analysis has determined that a mesocyclone was present. Note the obvious mesocyclone (denoted by a 1) in west central Illinois. Based on field surveys conducted by the National Weather Service, at least one tornado existed at this time.
Base Reflectivity Radar
The series of base reflectivity images (below) provide a close-up view of the Bounded Weak Echo Region (BWER) of the storm as it rumbled into Sangamon County and Springfield. The signature of the BWER is the "hook echo" that typically appears on a radar image of the southeast quadrant of a supercell, and marks the location of the storm's mesocyclone. The hook echo is an area of relatively light precipitation associated with the storm's updraft. Only a minority of storms possessing hook echoes produce tornadoes, but the evidence from the NWS storm survey confirms that a tornado existed during the time span of the images appearing below.
Doppler Velocity Radar
The Doppler velocity image (below) from 220Z on March 13th (820 PM CST on 3/12) shows the extreme rotation of the storm's mesocyclone near Curran, just to the southwest of Springfield. On Doppler velocity radar images, the areas appearing in green identify wind blowing toward the station while areas in red show wind blowing away from the radar site. It is the juxtaposition of both inbound and outbound wind that locates the area of rotation. Based upon NWS storm reports, a tornado was entering the southwest corner of Springfield at the time.
Because nearly all supercells produce some some sort of severe weather, an important facet of supercell forecasting is predicting the direction and speed of their movement so that severe weather watches and warnings can be appropriately placed. In the early stages of development, a supercell thunderstorm drifts along in accordance with the mean cloud-layer wind (the wind at 500mb is frequently used). Using the chart of 500mb observations (below) from 0Z on March 13, a supercell thunderstorm developing in the lower Great Plains would have been expected to move in a northeasterly direction.
The "30R75" Method
Not long after formation, instead of just "going with the flow", the dynamics affecting most supercell thunderstorms result in a modification of their course to the right of the mean cloud-layer wind. In 1976, R.A. Maddox formulated the 30R75 method as a relatively easy way to calculate the direction of supercell movement. The 30R75 method suggests that a supercell will move 30° to right of the mean wind direction at 75% of the mean wind speed. When originally developed, the 30R75 method relied upon wind observations at 850-, 700-, 500-, 300- and 200mb. For simplicity, the wind at 500mb can serve as the mean. Using the 500mb chart from 0Z on March 13th (above), the wind above the western border of Illinois was blowing in a direction estimated at 45° at a speed of sixty knots. Using the 30R75 method would result in a predicted storm movement of approximately 75° and a forward speed of forty-five knots.
The 30R75 method provides a simple means of calculating storm movement, but research has concluded that it is does not provide consistent results in a variety of vertical wind profiles, or when wind speeds are on the lighter side. Although the 30R75 method is not Galilean invarient, it provides a rough estimate of storm motion because most supercells in the United States move to the right and have a wind shear profile located within the upper-right quadrant of a hodograph. The motion of left moving supercells, although rare, cannot be calculated using the 30R75 method.
The Rasmussen Technique is Galilean invarient, and therefore, is a more sophisticated method of predicting storm movement. The Rasmussen Technique works in a wide variety of wind shear environments but requires wind observations and shear vectors to be plotted on a hodograph.
By using the Rasmussen Technique and observations from the Skew-T diagram from 00Z on March 13, 2006 (6 PM CST on March 12, 2006), we can estimate the motion of the supercells affecting central Illinois. Before turning to the hodograph, we must determine the wind speed and direction at the surface and at four kilometers above the surface (600mb will be used for this height). At 00Z, the surface winds were blowing from 140° at 12 knots. At 600mb, the wind was blowing at 54 knots from a direction of 240°. Using the Rasmussen Technique to plot these values on a hodograph, we can estimate storm motion in six simple steps:
Compared to the 30R75 Method, the Rasmussen Technique yielded a slightly different direction (240° versus 255°) but a dramatically slower forward speed (32 knots versus 45 knots). Based upon information from the NWS storm survey, the supercell responsible for the central Illinois tornadoes took an east northeasterly course, or approximately 247.5°.
Although both of these techniques provide a rough estimate of storm motion, the many variables -- both internal and external -- governing storm motion do not allow for consistently accurate results. The movement of a supercell over time, combined with the coarse distribution of radiosonde sites, makes it difficult to consistently obtain a timely and geographically significant sample of the mean wind speed and direction at a variety of atmospheric levels. Storm motion techniques that rely upon wind shear values at predetermined levels also have difficulty predicting the movement of very shallow supercells or storms over higher terrains. In addition, supercells approaching a synoptic boundary, such as a cold front or dryline, or dramatically different terrain, may abruptly shift course in response to changing atmospheric or orographic processes. In such situations, a storm motion technique that utilizes a hodograph alone will yield inaccurate results.
As mentioned above, the majority of supercell thunderstorms move to the right of the mean layer wind. This tendency is the result of differences in barometric pressure that develop as vertical wind shear interacts with the storm's updraft. In response to this vertical wind shear, an area of low pressure develops in the upper region of the downshear (right) side of the storm's updraft. In contrast, a region of relatively higher pressure forms on the other side of the updraft in the upshear portion of the storm. The region of low pressure on the right side of the updraft creates a favorable environment for propagation of the storm and, in response, it moves to the right.
Although rare, there are exceptions to the right-moving supercell. It is possible, in environments where directional wind shear is nominal, for an area of low pressure to form on both the left and right side of the storm's updraft. In some circumstances, the storm's downdraft splits the updraft in two: a contrary version that moves to the left of the mean wind and rotates anticyclonically, and a traditional right-moving model that rotates cyclonically. The future of the left-moving storm is guided by the nature of the storm's hodograph and the amount of wind shear.
Although far from perfect, predicting storm motion remains a valuable exercise and should improve as additional research is conducted.
Previous: Forecasting Supercells
© 2005-2006 Mark A. Thornton