The Siren Tornado: Destruction in Northern Wisconsin James Conrad, Department of Atmospheric and Oceanic Science, University of Wisconsin, Madison Wisconsin Abstract The Siren tornado is a somewhat classic case of the formation of a classic case supercell thunderstorm while also carrying some non-classical ideas of formation. A low level jet brings moisture into the affected area, with ambient directional and speed wind shear already located in the area. With daytime heating destabilizing an area with an already healthy mid-level elevated mixed layer, the setup for a severe thunderstorm outbreak was present, only requiring a force to break the boundary layer inversion cap. This forcing was a shortwave pressure trough located in the low to mid-levels of the atmosphere. A theory of tornadogenesis via local lake-induced dynamics is presented. What results is an F3 tornado, killing 2 and injuring over a dozen. The storm is classical in nature, demonstrating the capability of Doppler Radar to detect supercell characteristics, such as velocity coupling and hook echoes. It is because of these features that forecasters were able to protect the village of Siren, WI, from more deaths and injuries. ________________________________________________________________________ Introduction While many atmospheric scientists focus most of their tornado research on the central Plains of the United States, tornadoes occur in many other places of the county. 49 of the 50 states average at least 1 tornado per year; however every state has experienced at least one in history. On the northern edge of “tornado alley,” the state of Wisconsin has had its share of destructive tornadoes. Two (original Fujita scale) F5 tornadoes have occurred in the state, and numerous other F3 and F4 storms have affected the state. One such F3 tornado affected the northern portion of the state during the early hours of 19th June 2001. Hardest hit was the village of Siren, WI, a community of about 1000 (988 in the 2000 census). According to the National Weather Service Weather Forecasting Office (WFO) in Duluth, WI, “The most extensive damage was in a 6 block wide area in Siren, where numerous homes were leveled. There was also extensive structural damage to many buildings. The average width of the tornado was 1/8 to 1/4 mile, with the widest width being about a half mile. Preliminary indications are that the path length of the tornado was about 27 miles. Two people died as a direct result of the tornado, with another person killed indirectly after the tornado. In all, there were 16 injuries as a result of the tornado.” The question remains, what was the cause of this storm, and how was it so strong? The Siren tornado was forced by a multitude of synoptic and mesoscale forcings. A warm frontal passage placed the city in the “triple point” of a growing surface cyclone, and the storms were triggered by an approaching shortwave and the surface cyclone. This forcing was aided by a low level jet which advected moisture into the region behind the warm front throughout the day. Perhaps also aidingFigure 1 – A 3 panel plot for 12z on 18th June 2001 Eta model run. The upper left panel (a) is surface observations with high and low pressure centers and their central pressures noted, isobars every 4 millibars, upper right (b) is 900mb winds every 5 knots, and lower left (c) shows 850mb dewpoints every 10 degrees F starting at 30 degrees F. The low level jet produced warm moist air out ahead of the surface cyclone, which continued throughout the day. aided by a low level jet which advected moisture into the region behind the warm front throughout the day. Perhaps also aiding supercell formation is a dew point and temperature contrast caused by Lake Superior. Also important to the system is the presence of an elevated mixed layer, which adds to the instability of the atmosphere. All of these features work in unison to create destruction for the village of Siren on this fateful night. Data/Methods First, to get a grip on the overall synoptic pattern of the system, Unisys Weather synoptic maps were used. These graphics provided an initial view into the dynamics of the cyclone which produced the tornado, such as the center of low pressure and location of the warm and cold fronts. Surface features are provided that are author-created. Also used during the study is the ETA analysis and model run from the 18 June 2001 at 12z. This is used to create maps showing upper dynamics such as the low level jet, and low-to-mid-level height fields. Wind barb maps are created using GEMPAK and GARP. No data was available for the ETA model analysis on 19 June 2001 at 0z. Soundings are attained from the University of Wyoming sounding website and are from the Minneapolis (MPX) sounding site. Radar and satellite data and pictures are compliments of CIMSS.Figure 2 – The Minneapolis (MPX) sounding for 12z for 18th June 2001. Note the large inversion near the surface. Elements will become oriented correctly for a severe weather outbreak later in the day. Synoptic and Mesoscale Overview: Morning - Building Blocks The morning before the tornado began similar to other severe weather outbreak days. At 12z, a surface low with a central pressure of 1004mb located where the Missouri River forms the border between Nebraska and South Dakota was making its way east-northeast towards the region (Figure 1a, upper right). The pressure trough extending east from the center of the low's center is the warm front, which is located south of Siren, positioned across the Minneapolis-St. Paul metro area (MPX) towards the city of Eau Claire, WI (EAU). The cold front extends from the center of the low to the south-southwest, towards the southwest border region of Kansas and Nebraska. Cyclonic flow forced by the low would force air from the south into the region, where dew points were already in the high 50 to low 60 degrees F. Looking at 900mb winds, one can see a large maximum to the southwest of the affected region, with speeds in excess of 55 knots (Figure 1b, upper right). This jet maximum will move towards the affected region through the day, bringing more and more southerly air into the area. In association with this southerly air there is a copious amount of gulf moisture. ThisFigure 3 – Surface Observations at 0z on the 19th June 2001 and Conceptual Model for