During the afternoon of March 05, 1998, a significant severe thunderstorm event occurred over portions of southeast MS and southwest AL between Noon and 4 p.m. CST. In all, thirteen Severe Thunderstorm and two Tornado Warnings were issued by the National Weather Service (NWS) in Mobile,AL. One of the severe thunderstorms produced prolific severe hail fall as it moved east over Greene Co. MS between 1:38 p.m. CST and 2:30 p.m. CST (Damage Path). Dime to golfball-size hail fell in association with this storm as it moved between McClain and Leakesville,MS. Hail as large as baseballs fell near the the prison which is located just northwest of Leakesville. In addition, the storm was associated with two mesocyclonic circulations (Mesocyclone Tracks). The first circulation (M1) formed /evolved on the storm south flank and was the predominant circulation throughout south Greene Co. and was associated with the severe hail fall in Leakesville. The second circulation (M2) formed on the northwest storm flank and moved over the north portion of Greene Co. (forming just prior to M1's dissipation). The most severe hail damage occurred near Leakesville where the hail depth was reported to be 6 to 12 inches after the storm had passed. The elementary school in Leakesville,MS observed hail drifts as high as to the bottom of first story windows. Almost every vehicle that encountered the hail experienced damage. The severe hail also chipped paint on houses, dented house siding, stripped trees and destroyed satellite dishes. This is the largest known diameter hail to have ever fallen over Greene Co, MS since the NWS began keeping records in 1950. The following is the Leakesville storm damage path and a series of hail photos taken from the event (see Photos 1-4 below).
|Damage Path||Hail Photo 1||Hail Photo 2||Hail Photo 3||Hail Photo 4||Mesocyclone Tracks|
At 1200 UTC (6 a.m. CST), a zonal upper tropospheric flow was present over the southern US. The mid- and upper-tropospheric wind flow was diffluent in a zone that spanned east TX, AR and LA. Wind divergence was enhanced over a region that lie between a departing longwave trough off the Southeast Coast and an approaching upstream jet core (170 kt / 85 m/s) moving over Texas. The mid-troposphere was relatively cold (observed 500mb temperatures -15 to -20 deg(C)) and dry (observed 500mb dewpoint depressions 20-30 deg(C)) over the Central Gulf Coast Region in the wake of the aforementioned Southeast US trough. Wetbulb-zero temperature heights ranged from 9000-9500 ft agl in the regional airmass. An upstream 700mb shortwave trough was embedded in the zonal flow and was located over AR/LA. The Slidell,LA upper air sounding indicated unstable mid-tropospheric temperature lapses (18-20 deg(C)) with a CAPE value of approximately 1000 J/KG. From a vertical wind shear perspective, the 1200 UTC SIL Hodograph indicated the potential for intense, dynamically organized and rotating updrafts ( 0-2km storm-relative inflow = 152deg/18kt and 0-3km storm-relative helicity = 204 m2/s2 for 30R75). In the lower troposphere, a strong ridge was situated southeast of FL and was providing a moist S-SW return flow of air from east TX to the NW FL Panhandle (surface dewpointsas high as low 70Fs in E TX). One may note the nose of the low-level moisture axis advecting into southern MS/AL. By 1755 UTC (11:55 a.m. CST), deep convective activity had increased in coverage and intensity across the region. An E-W oriented line of thunderstorms was located between Tupelo,MS and Montgomery,AL and was moving south at 20 kt towards the AL/MS coast where the airmass was becoming increasingly unstable with time due to significant positive thermal and moisture advection.
|1500 UTC - 300mb||1200 UTC - 500mb||1200 UTC - 700mb||1500 UTC - 850mb||1200 UTC -RAOB||1200 UTC - HODOGRAPH||1200 UTC - SFC|
As mentioned in the discussion above, a vigorous upper troposperhic jet core (or 85 m/s at 12Z) was moving over Texas. In addition, a relatively weaker jet core (110kt or 55 m/s) was located over NC to the rear of the mid-tropospheric longwave trough which was departing the East Coast. Note the zone of relatively dry mid- and upper tropospheric air that lie along and just south of the major jet core axis. A vertical cross section taken from the 1500 UTC NCEP Meso-ETA model (valid 1500 UTC) between Lake Charles,LA northeast to Chattanoga,TN. This cross section was taken for the purpose of bisecting the W-E frontal zone that lie over MS/AL. The cross section revealed a potential ageostrophic frontal circulation that had developed over MS/AL ( ie.. the left exit region of the southern jet core had become coupled with the right entrance region of the northern jet core). Upward vertical motion was being enhanced along and north of the surface warm front (indicative of an intensifying frontal circulation). Upon comparison of the GOES-8 1400 UTC ( 8 a.m. CST) Meridian,MS sounding and the 1800 UTC (noon CST) GOES-8 Mobile,AL sounding, one can readily confirm continued modification / destabilization of the local airmass due to incident solar radiation that had been received during the late moring hours in the cloud free region over southeast MS/southwest AL ( see that CAPE values increased from near zero to around 700 J/KG as the morning low-level radiation inversion was eroded). In addition, GOES-8 satellite sounder data continued to confirm the existence of relatively cold mid-level temperatures (observed 1900Z GOES-8 500mb Temperatures -16 to -18 deg (C) ) very warm lower tropospheric air (observed 1900Z GOES-8 850mb Temperatures +8 to +10 deg(C) ) and regional wetbulb zero heights as low as 9000 ft agl which were all conducive to large hail production given ample updraft intensities within the local airmass. At 1846Z (or 1246 p.m. CST), the GOES-8 10.7 micron infrared imagery (IR) indicated a newly developed cluster of thunderstorms over southeast MS and southwest AL. This cluster of thunderstorms produced severe hail in both Creola and Saraland,AL by 1811 UTC (or 1211 p.m. CST) and would eventually merge with the line of storms moving southward over inland sections of south Alabama by 2100 UTC (or 4 p.m. CST ).
|1500Z - GOES-8||1846Z -GOES-8||1900Z - GOES-8||1400Z - GOES-8|
|1500Z - Vertical Cross Section||1800Z -||1900Z - GOES-8||1800Z - GOES-8|
The 1902 UTC (1:02 p.m. CST) IR satellite image revealed three distinct regions of relatively colder cloud tops ( -44 to -46 deg(C) ) that correlated well to the intense thunderstorm updrafts observed on the Mobile,AL WSR-88D. One region of colder tops was associated with a line of thunderstorms that was moving south across interior portions of south AL/MS. A second region of colder cloud tops was associated with a severe thunderstorm and was producing large hail near Creola,AL. The third region (located over NE Stone CO of southeast MS) was associated with a newly developed cluster of updrafts that would eventually become the Leakesville,MS severe hailstorm. Between 1902 and 1932 UTC( 1:02-1:32 p.m. CST) further cloud top cooling occurred (reaching the -52 to -53 deg(C) range ) and weak mid-level rotation was noted in a developing thunderstorm updraft that was located on the SE Perry/SW Greene CO Mississippi border. A Severe Thunderstorm Warning was issued at 1929 UTC (or 1:29 p.m. CST) by the NWS in Mobile,AL. By 1932 UTC, the storm was observed to possess the following characteristics : 1) a Weak Echo Region with an elevated mid-level reflectivity core of greater than 70 dBz 2) a Vertically Integrated Liquid (VIL) value of 59 kg/m2 3) a 100% Probability of Severe Hail (POSH) value (WSR-88D Hail Algorithm) with a 3 inch forecast severe hail diameter and a 4) maximum storm top of 38 kft agl. The storm was forecast to move directly over Leakesville,MS and was intesifying. As mentioned above, the Leakesville storm was associated with two mesocyclonic circulations (Mesocyclone Tracks). M1 formed /evolved on the storm south flank and was the predominant circulation as the storm moved over south Greene County. The M1 circulation center occupied the storm depth of 12-20 kft agl during its evolution. Between 1940 and 1950 UTC ( 1:30 - 1:40 p.m. CST), the Weak Echo Region became bounded (BWER) and the storm began to exhibit an appendage on the southwest flank while moving between McClain and Leakesville. The Severe Thunderstorm Warning for Greene Co. MS was upgraded to a Tornado Warning at 1952 UTC (or 1:52 p.m. CST) based on the aforementioned storm structural characteristics. Around this time, M1 was associated with a maximum Vr-shear value of .010 radians while extending as low as 7kft agl into the storm low-levels. Upon further evolving between 1955 and 2005 UTC , the BWER and storm top collapsed, overall storm rotation weakened, and the low-level appendage occluded. M2 (a weaker and more shallow mesocyclonic circulation) formed on the northwest storm flank just prior to M1's dissipation. Although neither mesocyclone circulation produced a tornado, around 2000 UTC (or 2 p.m. CST) extremely large diameter hail was observed just northwest of Leakesville,MS (near the prison). In addition, cloud tops reached a minimum value of -53 deg (C) at this time in association with the overshooting thunderstorm top located near Leakesville. Cloud top warming / updraft weakening was noted between 2015 UTC and 2045 UTC. A Tornado Warning was issued for Washington CO. AL while the storm was located over eastern Greene CO., MS. Although no tornado occurred with this storm, severe hail fell over southwest Washington CO, AL in association with this storm. The storm (and its associated mesocyclonic circulations) dramatically weakened upon entering south-central Washington CO. Alabama and no further warnings were issued.
|1902Z -GOES-8||1915Z -GOES-8||1932Z -GOES-8||1945Z -GOES-8|
|2002Z -GOES-8||2015Z -GOES-8||2032Z -GOES-8||2045Z -GOES-8|
|WSR-88D 0.5 deg REF -
|WSR-88D Composite Reflectivity -
|WSR-88D Storm Trends -
1907 - 1955 UTC
1900 - 2030 UTC
Special thanks to Gary Beeler (Warning Coordination Meteorologist) and Richard Davis (Meteorologist Intern) for conducting a detailed survey of the storm damage path. Photo credits go to Gary Beeler. Lastly, thanks to Eric Esbsensen (Meteorologist Forecaster) and Chris Davis (Meteorologist Student Trainee) for collecting some of the post-event data in order to prepare this post-event analysis.