National Weather Service United States Department of Commerce




During the early evening hours of May 8th, 2002, an isolated tornado event occurred within the Charleston, West Virginia County Warning Area (CWA). A weak, unorganized area of elevated showers moved in the vicinity of a warm frontal boundary late in the afternoon, during peak heating. An isolated thunderstorm rapidly developed and became surface based, as it interacted with the warm front, and eventually spawned a very short-lived tornado near Rio Grande in Gallia County, Ohio.


Synoptic and Mesoscale Pattern

The upper air pattern on this late spring afternoon featured a longwave trough across the intermountain west and a strong ridge over the Caribbean and southeastern United States, providing a broad west-southwesterly flow across the region. The 850mb (or about 3000 ft above the surface) map showed moist southwesterly flow and dewpoints in the 50 to 55 degree Fahrenheit range.

The early morning surface map (Figure 1) showed a nearly stationary frontal boundary, which extended from a surface low over Kansas, eastward through southern Illinois into eastern Kentucky. Surface temperatures were generally around 60 degrees north of the front and in the upper 60s to 70s south of the front. By late afternoon (Figure. 2) the frontal boundary had lifted north as a warm front and extended from north central Indiana through south central Ohio into southeast West Virginia. Temperatures south of the front had reached the upper 70s and surface dewpoints were in the upper 60s. The 00z Wilmington, OH upper air sounding indicated a very unstable atmosphere with Lifted Index values around -5 to -6. Storm relative helicity, an indication of how much shear is in the environment, was very high as well with values in excess of 250 m2/s2. Higher values of helicity (generally, around 150 m2/s2 or more) favor the development of mid-level rotation within thunderstorms. The warm front is suspected to have contributed to the enhanced environmental shear and played a significant role in the development of the Gallia County tornado.


Storm Development and Storm Interrogation

A cluster of weak showers and thunderstorms developed in the warm sector, south of the warm front, during the early afternoon hours on May 8th, 2002. Around 4:35 pm, the cluster of storms was approaching Jackson county Ohio (Figure 3) and did not show any severe thunderstorm characteristics. The warm front, at this time, was situated very close to Jackson and Rio Grand, Ohio. As the cluster moved closer to the front and began to interact with it around 5 pm, an isolated thunderstorm on the southern side of the cluster began to rapidly intensify as it moved through southern Jackson County (Figure 4). The storm relative motion (an indication of rotation in the wind field), or SRM, closest to the surface remained unorganized (Figure 5) but was showing signs of trying to organize. The storm of interest is believed to have crossed over the warm front between 5:30 pm and 6:00 pm. At 5:41 pm the KRLX Charleston, WV Doppler radar showed the thunderstorm crossing the Jackson and Gallia County line. A hook echo was evident (Figure 6), which is often associated with a rotating thunderstorm and indicates favorable conditions for tornado development. Strong convergence was present on the SRM image at 5:41 pm (Figure 7), represented by the 20-30 kt winds coming towards the radar (green), and the 20-30 kt winds going away from the radar (red). Notice that the maximum inbound and outbound values are not exactly against one another, which is indicative of convergence. When the two maximum inbound and outbound values are right next to one another, this indicates rotation and a possible tornado. The next radar scan at 5:46 pm (Figure 8) continued to show strong convergence in the lowest levels as the storm moved into the Rio Grande area. By 5:51 pm, a weak hook was still evident (Figure 9), although not as impressive as earlier, however, a weak tornado signature had developed (Figure 10) with maximum inbound winds of 30-40 kts and maximum outbound winds between 40-50 kts. The next radar scan at 5:56 pm showed the strongest tornado signature of the event (Figure 11), about 2-3 miles east of Rio Grande, with maximum inbound winds of 30-40 kts and maximum outbound winds in excess of 50 kts. Shortly after 6 pm, the Charleston, WV Doppler radar showed the storm moving towards the Ohio River (Figure 12) with a weaker and much more unorganized SRM image, indicating very weak rotation.


Event Summary

A storm survey that was conducted by the National Weather Service determined that a tornado briefly touched down between 6:06 pm and 6:07 pm along Adamsville Road east of Rio Grande. The tornado was on the ground for a very short time, but produced F2 damage according to the Fujita Scale with maximum estimated wind speeds of 120-125 mph. The tornado path was approximately 3 miles in length and 75 to 100 yards wide (Figure 13). Damage along Adamsville Rd included a roof ripped off a home, two trailers and a pole barn completely destroyed. The tornado then struck both the east bound and west bound rest areas along U.S. route 35 about 3 miles east of Rio Grande. The west bound rest area on the north side of U.S. route 35 suffered the most damage. The rest area building was severely damaged, at least two tractor trailer trucks were flipped over and several other parked vehicles were damaged. As the tornado proceeded northeast beyond the rest area, several other modular homes and trailers were destroyed or severely damaged. Fortunately, there were no casualties, but there were three people injured at the rest stop who were treated and released promptly from the hospital. The total damage from the tornado was estimated at 3.5 million dollars.


Additional images of the damage caused by the tornado can be viewed by clicking here.