National Weather Service United States Department of Commerce


On the evening of May 19th, 2002, a long-lived supercell thunderstorm tracked across eastern Washington. The storm moved across a sparsely populated area after sunset, so no damage was reported. But the supercell was impressive from a structure and evolution standpoint.


On the afternoon of May 19th, deep south-southeasterly flow existed over most of the Pacific Northwest with a large negative tilt trough offshore. This synoptic pattern is well known for being a favorable pattern for severe weather in the eastern Washington and northern Idaho area. This was well documented by Evenson and Johns in 1995.

Examination of the model forecasts for the afternoon (00Z 20 May) indicated good instability (Fig 1) over the Idaho panhandle with strong 0-6km shear (40-50kts). This pointed the forecasters to the possibility of organized convection over the panhandle. Low-level (0-3km) shear was also strong, further suggesting the possibility of organized lines. Over eastern Washington, stronger shear existed. But the drier low-levels minimized the available CAPE, with a large area of CIN present that would be difficult to overcome with daytime heating. An Eta forecast sounding from Moses Lake in the Columbia Basin (Fig 2) showed this less than impressive sounding..

Convection developed over the central Idaho mountains and moved north into the panhandle. An organized west-east line developed and produced severe wind damage in several locations. By the end of the event, the convective line had moved into Canada (Fig 3). Meanwhile, a lone thunderstorm developed over the mountains of northeast Oregon (see Fig 3) and moved towards the north-northwest into the low elevations of central Washington. While this area was unfavorable for initiating convection due to the strong cap, the thunderstorm that moved into this region was elevated and did not initially rely on the boundary layer.

As the storm continued to develop, it appears that the downdrafts generated by the storm were able to mix the low level air and thus remove the cap. The forecast straight-line hodograph (Fig 2) indicated that significant rotation was not to be expected unless the storm could move considerably left or right of the mean wind. At 0437Z some rotation was noted (Fig 4) but it was only in the weak mesocyclone category. Also note that the rotation was aloft, 16-21kft AGL. Several times during it's life the storm showed the classic structure of new updraft development. At 0449Z a new updraft is clearly evident on the southeast flank of the storm (Fig 5). Just 17 minutes later, a strong reflectivity gradient formed on the northeast side of the storm (Fig 6). This is an atypical region of the storm for this feature, but the low-level inflow was from the east-northeast as shown in the VWP (Fig 7). The storm reached full maturity at 0518Z (Fig 8) with 65 dbZ echoes present above 12kft AGL.


Evenson, E. C. and R. H. Johns: 1995: Some Climatological and Synoptic Aspects of Severe Weather Development in the Northwestern United States. Natl. Wea. Dig., 20, 34-50. 


By Ron Miller