Kenneth L. Pryor
Office of Research and Applications (NOAA/NESDIS)
Camp Springs, MD
Date Submitted:
Abstract
A case study using satellite imagery, satellite derived products, and radar imagery was conducted of the development and evolution of a severe squall line that occurred on the evening of 24 October 2001. One of the most widespread convective high wind events of the season, this squall line formed over the central Mississippi Valley then moved rapidly eastward to the Appalachian Mountains during the next 17 hours. Damaging wind reports were extensive (421 reports) with this squall line; thus, the 24 October 2001 event qualifies to be considered a derecho. Radar observations of the evolving squall line show signatures often correlated with damaging surface winds, including bow echoes, weak echo channels and reflectivity troughs. Synoptic conditions under which the 24 October 2001 event occurred were typical for the development of "non warm season" derechos: strong forcing by a vigorous, negatively-tilted, upper- level trough over the central United States and strong unidirectional vertical wind shear aloft over the derecho genesis area. Similarities to warm-season derecho conditions included convective instability and large amounts of buoyancy over the derecho genesis area. In addition, there was significant low level moisture present overlain by a cool, dry air layer in the middle troposphere. Geostationary Operational Enivronmental Satellite (GOES) Sounder-derived microburst products were also useful in assessing the potential for convective downbursts. The sounder-derived products used for this purpose were the Wind Index (WINDEX) for estimating maximum possible convective wind gusts and the Theta-e Differential (TED) for indicating the potential for wet microbursts. GOES soundings were used in the analysis to display the sounding profiles that were favorable for the formation of wet microbursts.
1. Introduction
During the afternoon and evening of 24 October 2001, a squall line progressed from the Mississippi Valley east to the Appalachian Mountains. The squall line was triggered by strong surface convergence along a cold front and resulted in over 400 reports of wind damage as well as two deaths and 20 injuries. There were numerous reports of damage rating F2 on the Fujita scale (Fujita 1971) to include downed trees and powerlines, rail cars blown over, broken windows, roofs blown off homes and overturned trailers. In addition, there were several reports of damage rating as high as F3 on the Fujita scale to include overturned trucks. Based on the widespread wind damage that occurred with this squall line, this system can be considered a derecho. A derecho is defined as an extratropical convective weather system that produces a family of downburst clusters (Johns and Hirt 1983). The downburst is defined as a strong convective downdraft that induces an outburst of damaging winds on or near the earth's surface (Fujita and Wakimoto 1983). Downbursts are classified into two types based on the temporal and spatial scale of the event. A macroburst is a large downburst with 4 km or larger outflow size with damaging wind persisting for 5 to 20 minutes. A microburst is defined as a small downburst, less than 4 km in outflow diameter, with peak winds persisting for 2 to 5 minutes (Fujita and Wakimoto 1983). Maximum winds were recorded as high as 77 kt (40 m s-1) with a downburst at South Bend, IN. Most measured wind reports associated with this event were in the 40 to 70 kt (21 to 36 m s-1) range in the central and eastern United States. Geostationary Operational Enivronmental Satellite (GOES) Sounder-derived microburst products were useful in assessing the potential for convective downbursts. For this case study, data was collected on the sounder-derived products Wind Index (WINDEX) and Theta-e Differential (TED). The WINDEX indicates the speed of maximum potential convective wind gusts at the surface (McCann 1994). TED is defined as the difference between maximum theta-e at or just above the surface and minimum theta-e in the middle troposphere. TED is an indicator of the potential for wet microbursts where a value of 20 or higher indicates a high risk (Atkins and Wakimoto 1991). The reader is referred to Ellrod et al 2000 for a complete description of GOES sounder-derived microburst products. Table 1 compares measured wind speeds during this event, obtained from Storm Prediction Center (SPC) storm reports and METAR observations, to GOES Wind Index (WINDEX) and Theta-e Differential (TED).
|
Table 1. Measured Wind Speed vs. GOES WINDEX/TED | |||
Location |
Measured |
WINDEX |
TED |
St. Charles, MO |
58 | 39 | 17 |
| St. Louis, MO | 44 | 39 | 17 |
| Lincoln, IL | 53 | 40 | 17 |
| Cahokia, IL | 45 | 39 | 17 |
| Decatur, IL | 41 | 40 | 17 |
| Lawrenceville, IL | 53 | 44 | 26 |
| South Bend, IN | 77 | 37 | 26 |
| Groveport, OH | 50 | 40 | 20 |
| Columbus, OH | 43 | 40 | 20 |
| Lancaster, OH | 44 | 41 | 20 |
| Columbus, MS | 61 | 43 | 27 |
| Zanesville, OH | 40 | 32 | 20 |
2. Synoptic Conditons associated with a "Non Warm Season" Derecho
The derecho evolved in an environment that was similar to the environment in which another "non warm season" derecho developed on 9 April 1991 (Duke and Rogash 1992). GOES visible imagery, Figure 1, displayed a vigorous, negatively-tilted, upper-level trough located over the central United States that moved eastward from the central Plains to the Mississippi Valley during the afternoon and evening of 24 October. The derecho genesis area (central Mississippi Valley) was located in a part of the circulation often considered favorable for the development of severe weather: winds aloft were strong and diffluent just east of the approaching trough. There was strong low-level southerly flow with low -level warm and moist air advection ahead of a rapidly moving cold front. Dew points were abnormally high for the season ahead of the cold front, ranging from 63 to 70F (17 to 21C). Convective instability was present over the derecho genesis area. Cool, dry air was present in the middle troposphere, with warm, moist air at lower levels, as displayed on the following soundings:
As indicated in the soundings displayed above, lifted indices were -5 to -7 with convective available potential energy (CAPE) ranging from 1184 at Evansville, IN to 1686 at St. Louis, MO. The soundings also indicate a relatively large negative gradient of theta-e in the vertical. Surface theta-e values ranged from 335 K to 343 K while mid-tropospheric values of theta-e ranged from 315 K to 323 K. It has been noted that a TED greater than or equal to 20 is conducive to strong convection and downbursts (Atkins and Wakimoto 1991). Large amounts of buoyancy and appreciable vertical shear at low levels favored the development of a long-lived squall line. Vertical wind shear aloft was unidirectional. There was strong vertical wind speed shear with no significant change in wind direction with height. In this case, the combination of strong convergence and low-level forcing was the primary trigger for thunderstorm activity.
3. Derecho Evolution
1600 UTC central Mississippi Valley radar composite (NCAR/RAP), Figure 3a, displayed an area of isolated single cell and multicell thunderstorms extending from central Illinois southwest to southern Missouri. GOES Wind Index (WINDEX) values at 1600 UTC, Figure 4a, in the derecho genesis area, ranged from 40 to 45 kt. By 1800 UTC, the area of thunderstorms had evolved into a squall line over the central Mississippi Valley. Figure 3b displays a northeast to southwest oriented line of thunderstorms with bow echoes developing over central Illinois and eastern Missouri. Weak echo channels were apparent on the back side of the bow echoes, signifying the presence of downburst winds (Przybylinski and Gery 1983). The weak echo channel identifies the location of a rear-inflow jet which feeds dry, higher momentum air into the downdraft, enhancing the strength of the resulting outflow at the surface through vertical momentum transport and increased evaporation (Weisman 2001). Wind gusts of up to 45 kt occurred in the St. Louis area while a gust of 53 kt was recorded at Lincoln, IL. The highest wind gusts occurred near the tight reflectivity gradient at the leading edge of the bow echoes. 1800 UTC WINDEX, Figure 4b, ranged from 36 kt over northern Indiana to 50 kt over southeastern Illinois. By 2100 UTC, Figure 3c indicated that bow echoes were embedded in the squall line over north-central Indiana, southern Illinois and southeastern Missouri. The bow echo over northern Indiana, just south of South Bend, had taken the shape of a spearhead and was accompanied by a weak echo channel located along the center axis of downburst flow. The highest downburst wind gusts would be expected to occur near the weak echo channel. Consequently, a wind gust of 77 kt was recorded at 2104 UTC at South Bend, IN as the squall line moved east at a speed of 40 kt. Translational speed of the squall line was determined by comparing the geographical location of the squall line axis displayed in a composite radar image to the location of the squall line displayed in a radar image one hour prior.
As the squall line continued to move east at a speed of approximately 40 kt., 0000 UTC WINDEX, Figure 4c, over central Ohio and northeastern Mississippi ranged from 40 to 43 kt. By 0100 UTC, Figure 3d indicated that the squall line extended from eastern Michigan southwest to northern Mississippi. Bow echoes were developing along the line over southwestern Ohio, Kentucky, and southwestern Tennessee. At 0200 UTC, Figure 3e displayed a well-defined bow echo embedded in the squall line over central Ohio, just west of Columbus. Again, a weak echo channel was apparent on the back side of the bow as downburst winds were intensifying (Przybylinski and Gery 1983). Maximum wind gusts of up to 50 kt. in the Columbus, OH area occurred as expected: adjacent to the tight reflectivity gradient at the leading edge of the bow echo. A reflectivity trough, an area of lower reflectivity between strong leading echoes and trailing echoes just west of the weak echo channel, indicated that dry air was accelerating in to the system from the its rear. This overtaking air can contribute to wind gusts along the leading edge of the squall line (Duke and Rogash 1992).
By 0300 UTC, Figure 3f indicated that the squall line extended from Lake Erie southwest across Lousiana. A well defined bow echo was located on the line near Columbus, MS. A maximum wind gust of 61 kt. was observed at Columbus, MS. After producing some weaker downbursts, the squall line eventually dissipated over the Appalachian Mountains between 0600 and 0900 UTC 25 October 2001.
4. Downburst Generation Physical Processes
Thunderstorm development during this event was the result of strong convergence along and ahead of a rapidly moving cold front. Forcing was concentrated along the thunderstorm gust front where there was an abrupt shift from southerly winds at 10 kt ahead of the gust front to westerly winds at 40 kt just behind the gust front. The strong convergence produced by this velocity field allowed convection to propogate along the thunderstorm outflow boundary.
The generation of downbursts associated with this squall line was attributable to several factors also noted in the 9 April 1991 derecho (Duke and Rogash 1992). Strong updrafts penetrated the elevated dry layer. The dry air cooled evaporatively as it entrained in to the moist updrafts. Cooling also took place where the precipitation evaporated as it fell through the dry air. The chilled air became negatively buoyant and accelerated downward. Strong instability was present in the air mass ahead of the cold front: lower levels of the atmosphere were warm and moist while the higher levels were significantly cooler and drier. Downward transport of higher momentum possessed by winds in the mid-troposphere was apparent. Parcels in the elevated dry layer conserved horizontal speeds as they became negatively buoyant and descended to the surface. The wind direction associated with the downbursts indicated a contribution from downward momentum transfer. Maximum wind gusts had a large westerly component: directions ranging from 250 to 320 degrees while dry layer wind directions ranged from 240 to 260 degrees. High radar reflectivities (>55 dBz) were associated with the downburst wind gusts. This suggests that precipitation drag or water loading was a factor in the generation of downbursts.
5. Summary and conclusions
Synoptic conditions associated with the 24 October 2001 derecho were typical for "non warm season" derecho events. Strong forcing by a vigorous upper-level trough and a rapidly moving cold-front served as a trigger for the development of thunderstorm activity. In addition, winds aloft were strong and diffluent with significant vertical unidirectional wind shear over the derecho genesis area. However, there were some similarites to the conditions associated with warm-season derechos. There were strong instability and buoyancy over and east of the derecho genesis area. Significant low-level moisture was present along and ahead of the cold front overlain by a cool, dry air layer in the middle troposphere. The mid-tropospheric dry layer provided a source for evaporative cooling and the formation of strong downdrafts. Strong winds in the dry layer supplied momentum that could have been transferred to the surface by the intense downdrafts (Duke et al 1992). The vertical unidirectional wind shear present over and east of the derecho genesis area promoted the development of the area of thunderstorms into a squall line. Since this derecho convective system consisted of an extensive squall line in which downburst activity occurred with a series of bow echoes that moved along the line, this squall line pattern could be referred to as a serial derecho (Johns and Hirt 1987). As typical with serial derechos, the strongest downburst activity was observed toward the north end of the squall line, especially over northern Indiana.
The underestimation of convective wind gusts by the WINDEX was most likely attributable to the rapid translational motion of the squall line during the majority of its duration. Accurate forecasting of maximum convective wind gusts can be accomplished by extrapolating WINDEX values and then adding the translational speed of storms to the WINDEX. For example, 1800 UTC WINDEX indicated a value of 36 kt over northern Indiana. Adding storm motion (40 kt) to the WINDEX yielded a value of 76 kt, one kt less than the actual downburst wind gust speed (77 kt) recorded at South Bend, IN at 2104 UTC. GOES soundings and radiosonde observations were instrumental in displaying the sounding profiles that were favorable for the formation of wet microbursts. In addition, the use of radar imagery was important to display the signatures which were associated with strong convection and downburst generation (i.e. bow echoes, weak echo channels, reflectivity troughs, etc.). This derecho event demonstrated the importance of accounting for storm motion and downward advection of momentum when using the WINDEX to forecast maximum possible convective wind gusts.
6. References
Atkins, N.T., and R.M. Wakimoto, 1991: Wet microburst activity over the southeastern United States: Implications for forecasting. Wea. Forecasting, 6, 470-482.
Duke, J.W., and J. Rogash, 1992: Multiscale Review of the Development and Early Evolution of the 9 April 1991 Derecho. Wea. Forecasting, 12, 623-635.
Ellrod, G.P., J.P. Nelson, M.R. Witiw, L. Bottos, and W.P. Roeder, 2000: Experimental GOES Sounder Products for the Assessment of Downburst Potential. Wea. Forecasting, 15, 527-542.
Fujita, T.T., 1971: Proposed characterization of tornadoes and hurricanes by area and intensity. SMRP Research Paper 91, University of Chicago, 42 pp.
Fujita, T.T., 1979: Objective, operation, and results of Project NIMROD. Preprints, 11th Conf. on Severe Local Storms, Kansas City, MO, Amer. Meteor. Soc., 259-266.
Fujita, T.T., and R.M. Wakimoto, 1983: Microbursts in JAWS depicted by Doppler radars, PAM and aerial photographs. Preprints, 21st Conf. on Radar Meteorology, Edmonton, Amer. Meteor. Soc., 638-645.
Johns, R.H. and W.D. Hirt, 1983: The derecho: a severe weather producing convective system. Preprints, 13th Conference on Severe Local Storms, Tulsa, OK, Amer. Meteor. Soc., 178-181.
Johns, R.H. and W.D. Hirt, 1987: Derechos: Widespread Convectively Induced Windstorms. Wea. Forecasting, 2, 32-49.
McCann, D.W., 1994: WINDEX-A new index for forecasting microburst potential. Wea. Forecasting, 9, 532-541.
Przybylinski, R.W., and W.J. Gery, 1983: The reliability of the bow echo as an important severe weather signature. Preprints, 13th Conf. on Severe Local Storms, Tulsa, OK, Amer. Meteor. Soc., 270-273.
Weisman, M.L., 2001: Bow Echoes: A Tribute to T. T. Fujita. Bull. Amer. Meteor. Soc., 82, 97-116.
Acknowledgements
The author thanks Jamie Daniels (NESDIS/Forecast Products Development Team) and Raytheon contractors for providing GOES sounding retrievals displayed in this paper. Jackson, MS (JAN) radiosonde observation was provided by UNISYS Corporation web site (URL: http://weather.unisys.com). Radar imagery (NIDS/NEXRAD) was provided by Peter Neilley, National Center for Atmospheric Research, Research Applications Program (NCAR/RAP), via the NEXRAD Data Archives Viewer: http://www.rap.ucar.edu/staff/pneilley/NIDS_archives.html.