Generated using V3.0 of the official AMS LATEX template–journal page layout FOR AUTHOR USE ONLY, NOT FORSUBMISSION!Dianosis of Tornadic Supercell Formation over Southern WisconsinMichael A. Balliett∗University of Wisconsin-Madison, Madison, WisconsinABSTRACTOn August 18th, 2005 a tornadic outbreak occured in central and southeasternWisconsin. Prelinimary reports are dominated by tornadoes. There were also afew large hail reports, but these were few and far between, and wind reports werenon-existant. Radar from the event showed storms with classic supercell shapeand movement. Of particular interest are two heavy precipitating supercells thatformed in along and ahead of the cold front in southern Wisconsin. It was foundthat these storms had a better chance of producing tornadoes because of their lowcloud condensation and level of free convection that occured in an environment withhigh low level shear. These two mesoscale properties of the storms lead to a strongmesocyclone that formed near the surface. With the meso so close to the ground,any vocticity formed below the meso could easily be transformed into a tornado.1. IntroductionLarge tornado outbreaks are rare in Wisconsin.In addition, one rarely gets that many tornado e swhen an outbreak does occur. This is evidencedsince the record one day total of tornadoes in Wis-consin is 26 tornadoes. This outbreak occurred onthe afternoon of August the 18th. In this case, theNational Weather Service only issued a slight riskof severe weather. This risk was warranted, butthe storms that did form had plenty of low levelshear, which could be transformed into a tornado.These tornadoes were not weak also of the 27 tor-nadoes formed, 16 of them were rated F1 or aboveby the national weather service. The strongest ofwhich was an F3 that struck Stoughton, WI. View-ing a plot of the paths of the tornadoes, one willnotice two principal areas in which the tornadoesformed. Either, the tornado occurred along a lineextending from just east of Tomah, WI east north-eastward toward Green Bay. Then there is as sec-ond line that starts in eastern Vernon county andextends east-southeastward into south west Jef-ferson county. The southern line of tornadoes hadmore tornadoes (19) and also had stronger torna-does (all of the F2 and F3 tornadoes). This differ-ence in strength can be attributed to the type ofstorms that created the event. The southern tracktornadoes were created by a strong supercell thatgrew out of a convective line. The northern tracktornadoes grew out of a vortex that formed on thenorth side of a squall line.The supercell that formed the tornadoes insouthern Wisconsin formed more and stronger tor-nadoes, because supercells are more condusive tocreating tornadoes. In a typical supercell, verti-cal shear is tilted in the updraft to produce anupdraft that rotates, otherwise known as a meso-cyclone, see Figure 1. Many effects result fromthis rotation. Most importantly is that the rota-tion of the updraft creates an inertial wall whichprevents intrusion of dry air into the updraft. The1rotation also spins out heavier particles in the up-draft. These particles such as rain, hail and gaupelform in an intense band of precipitation as seen onradar just to the outside of the updraft, see Fig-ure 8. Typically, the heavier particles only getpartially wrapped around the mesocyclone and onradar, there appears a hook echo.The water loading outside of the mesocyclonealso helps to facilitate the development of two prin-cipal downdrafts. The forward flanking downdraftis formed in association with dry air from the backside of the supercell working its way underneaththe cell, and then having rain fall into the dry aircools it off and moistens it. This forces the airto fall to the surface and form a stationary frontout in head of the storm. The rear flanking downdraft is created by the upper level winds beingforced downward when they hit the updraft. Thenas they fall, they fall into the rain that has beenevacuated from the updraft cooling and moisten-ing the air as it falls. This air too falls to theground and spreads out forming a rear-flankingoutflow. The sometimes on the north side of therear flanking down draft, there can be cyclonicvorticity that can be kicked up due to horizontalspeed shear. If this can become move under themesocyclone, then this could provide an excellentopportunity for a tornado to develop. Further-more, this outflow can actually get out ahead ofthe updraft combining with the forward flankingoutflow boundary. This in effect kills the updraftas it is now isolated from the warm moist air thatwas feeding it.The life of the storm is renewed by the cyclonicshear though. As the storm occuludes there is twodynamic lows that form to the left and right ofthe original updraft. One will have cyclonic shearand the other will have anticyclonic shear. Whichone is cyclonic and which one anticyclonic dependson the flow in the vertical. The updraft with cy-clonic flow will have chance to survive and becomethe new supercell, while the other will quickly die.The most typical arrangement of cells is to havethe cyclonic flow to the right and the anticyclonicflow to the left. This means that most typicallythe right storm is favored for development, giv-ing the storm in general the appearance of movingright.2. DataMany different media performed the analysis ofthis storm. For an overview of the meso- and syn-optic scales setup, Storms Prediction Center eventanalysis was used. This included mesoscale discus-sions that were issued throughout the day, the se-vere weather outlooks that were updated, and thestorm reports. Initial analysis of infrared satel-lite loops and radar loops was also done, using theStorms Prediction Center website. Through these,a more specific period was found that best corre-lated with the time that the event was set up andthe event occurred, August 18th 12Z to August19th 04Z. In addition, surface data from weatherstations across the United Stated was availablehourly and upper air data was available for ev-ery 12 hours across the United States. For thetimes between the sounding 00Z and 12Z hours,i.e. the times soundings were launched, the etamodel that was initialized with the sounding andsurface observations was used. General Meteorol-ogy Package (GeMPak) was used to image modeland observed data along with model soundings. Inaddition, soundings were retrieved from the Uni-versity of Wyoming achieve. In addition, skew-tand hodograph