June 11, 2001 Storm Analysis
This storm began to develop in extreme northwestern Wisconsin as cells merged into one squall line. The first damage report was received at 5:10pm from Siren in Washburn County. There was a 60-mph wind gust and a large branch over a highway. The storm raced southeasterly very quickly and covered lots of ground. At 7:00pm, the storm still hadn’t lost its punch as it screamed through Eau Claire with near 70-mph gusts, ripping up trees and power lines. The first radar picture I printed was at 8:45pm, just as the storm was beginning to bow out into an echo shape, soon to become a derecho. A derecho is a widespread convectively induced windstorm. At 9:00pm, the first warnings in Southeast Wisconsin were issued for Dodge and Fond du Lac Counties and were severe thunderstorm warnings.
My mentor, John Eise, gave me some radar images from the National Weather Service. The 9:00pm image showed that the echo was bowing out near Dodge and Fond du Lac Counties. This meant that there were strong winds in the bowing part of the storm. At 9:29pm, there were several indications on radar that there may have been tornadoes on the ground in Dodge and Fond du Lac Counties. A Tornadic Vortex Signature appeared north of Fox Lake in Dodge County. A shear marker indicated a mesocyclone in the same area north of Fox Lake. Later, it was confirmed the tornado in Dodge County actually happened at that exact time. There were no warnings out when this actually occurred. At 9:40pm, the National Weather Service issued severe thunderstorm warnings for Sheboygan, Washington, Ozaukee, Jefferson, and Waukesha Counties. Also at this time there was a very distinct bookend vortex or a rotating head at the northern end of the derecho. It looked very much like a low-pressure area, only smaller. The radar also showed rear inflow notches, which are areas of higher winds, on the backside of the derecho.
Even though the National Weather Service did not issue a tornado warning earlier, they did issue one at 9:53pm for Dodge and Washington Counties because they saw something that looked very much like a tornado on their radar. This tornado was unconfirmed. At 9:50pm, the derecho was making an impact on or threatening the following cities: Sheboygan, Two Rivers, West Bend, Waukesha, Jefferson, and Madison. It looked as though it would be knocking on Milwaukee’s doorstep in about half an hour.
At 10:00pm, the National Weather Service issued severe thunderstorm warnings far in advance of the derecho in Milwaukee, Racine, Kenosha, and Walworth Counties. At 10:06pm, tornado warnings began in Milwaukee, Ozaukee, Waukesha and Washington Counties. Sirens also blared in Kenosha, Racine, and Walworth Counties due to high wind speeds advancing towards these areas. However, no tornado warnings were issued. At this time, I also wrote in my storm timeline that I thought the southern edges of the storm were weakening and I was correct.
At 10:39pm the derecho was already in Racine and Lake Geneva, but not as intense as before. When the storm finally got to Kenosha, the very corner of the state, it had weakened so much that it only managed a 40-mph wind gust, well short of severe classification. All warnings were cancelled at 11:01pm.
Many people know what’s happening in the lowest levels of the atmosphere, such as the current temperature. However, they do not realize what’s going on in the upper levels of the atmosphere. To forecast thunderstorms, you have to look in the upper atmosphere first. BUFKIT is a program used by the National Weather Service to analyze and forecast various conditions in the upper atmosphere, such as CAPE, CINS, LFC, and Helicity.
CAPE stands for Convective Available Potential Energy. It is basically a measure of how much energy is available for convection. Convection in meteorology usually refers to vertical motions. Therefore, it is related to the maximum wind speed in an updraft. A good CAPE level is in excess 1500 joules per kilogram. An excellent level would exceed 2000 j/k. Extreme amounts can reach 4000 j/k.
CINS stands for convective inhibition. CINS is a measure of the amount of energy needed in order to start convection. In other words, the lower the CINS level is, the easier it is for thunderstorms to form. A good CINS level is lower than 30 joules per kilogram. An excellent CINS level is lower than 15 joules per kilogram. These values reflect the strength of the cap. The cap or capping inversion, is a layer of warm air aloft, which stops or delays thunderstorm development. This is because when a warm, moist air parcel is rising it needs to be warmer than the surrounding air. When it reaches the cap the warm air inhibits the air parcel from rising any higher and therefore, no thunderstorms can develop.
The LFC, or level of free convection, should be low in order for thunderstorm development to occur, usually less than 7500 feet. This is because when moist, warm air is rising it has to fight upwards against warmer air or a weak cap until it reaches the LFC, where it will rise uninhibited. The lower the LFC is, the less an air parcel has to fight against warmer air. If the LFC is too high, the air parcel will lose much of its energy by the time it get to that level.
Helicity is not a key element in normal thunderstorm development. It instead tells us how favorable the conditions are for atmospheric rotation. The higher the helicity, the better chance for the formation and development of mesocyclones and tornadoes. Helicity is basically showing you the amount of vertical wind shear and turning, or vorticity, in the atmosphere. Vertical wind shear means winds move in different directions at different levels of the atmosphere. A helicity level that is favorable for rotation is anything near or above 150 meters per second.
For horizontal rotation to form, there has to be vertical wind shear and the winds must shift in a counterclockwise direction as you go up in the atmosphere. It is very much like rolling a pencil between your hands. Once a storm forms, the horizontal rotation is in place and begins to suck in warm, humid air to feed the thunderstorm. This strong inflow causes the gust front to be held closer to the storm, thus increasing chances for a long life span. The strong inflow also helps create strong downdrafts and strong updrafts. The strong winds of the updraft and downdraft help tilt the rotation into a vertical position increasing probabilities of a mesocyclone or tornado.
The day the derecho occurred, June 11th, CAPE amounts were particularly high from 4:00pm to 11:00pm. This meant there was a good chance for severe weather to occur because higher values of CAPE mean a greater potential for severe weather. A good CAPE level is above 1500. An excellent level is above 2000. That day, the CAPE level ranged from 2187 j/k to 2743 j/k for most of the day, which indicates a good chance for severe weather. Also higher values indicate stronger updrafts, which tend to occur in severe thunderstorms. There is a formula that gives you a rough idea of how fast the updraft could be. First, you multiply the square root of CAPE by two. This answer gives you the speed in meters per second. If you want to convert it into knots you multiply by two again. Then if you want to convert it to miles per hour you multiply by 1.15. If you look at the 11:00pm level of CAPE, you’ll see the maximum updraft strength was 225 miles per hour.
Even though we had high levels of CAPE, they did not generate severe weather in Milwaukee prior to 11:00pm because the CINS levels were too high. The CINS ranged from 88 to 183 until they improved to 43 at 11:00pm. This means there was a relatively strong cap in place until it was weakened quite a bit at 11:00pm. Moist, warm, humid air could then rise more easily into the updraft of the storm to keep it going. The LFC level was low, about 4,500 feet, which meant the moist air did not have to be pushed as high. Also HELICITY levels were moderately high, at 163, which meant that rotation was favorable. This could be one of the reasons that there were tornado warnings in counties near Milwaukee. There was a confirmed F1 tornado in Dodge County. F1 tornadoes are considered weak tornadoes with winds ranging from 73-112mph. This one lasted twenty-two minutes.
By the time the derecho affected my house and the Kenosha area, dry and cool air had settled in. Thunderstorms need moist, warm air to keep them going. This is because moisture is needed for condensation or the formation of clouds. With moisture, the cumulus clouds grow higher or until they reach the cumulonimbus stage. Moisture is also needed for precipitation.
Starting at 10:00am the day of the derecho, June 11th, the winds in the Kenosha area shifted to a southeasterly direction, bringing them off the lake. This brought in drier air from over Lake Michigan. Before 10:00am the dew point had risen to near 70, but after 10:00am the dew point dropped gradually to 60. This not only brought the dew point down, but it also had a cooling effect and the temperatures gradually decreased. Due to the lake breeze, cool, drier air had settled in just off the lakeshore (I live 2 miles away from Lake Michigan). The conditions were becoming less and less favorable from noon on for thunderstorm development in Kenosha. When the derecho met the cooler, drier air in Racine County, just north of Kenosha, it started to dissipate.
Another interesting point is that the winds shifted right before the gust front hit. The gust front is a boundary between warm, humid air before the storm and the cool downdraft of the thunderstorm, bringing with it the first gusts before the actual thunderstorm hits. The winds shifted from south-southeast off the lake to north-northwest in seven minutes! The next minute, 10:45pm, was when the gust front hit bringing the first gust of 19 miles per hour. The gust front hit 11 minutes before it started raining. I like to think of the gust front as a warning to get inside because it usually hits about 10 minutes before the rain hits. So when it gets gusty you know its time to get inside provided the sky is cloudy and it looks like rain.
The Kenosha area mainly received heavy rainfall with the thunderstorm and the subsequent shower bands. It began raining at 10:56pm. In the first 10 minutes we received .23 inches of rain. This meant the rain rate was 1.38 inches per hour. This is pretty heavy rain. It stopped raining at 11:15pm. This meant it was raining for 19 minutes. After the initial storm, there were off and on shower bands well past midnight.
When the weakened derecho got to my house, it had become very depleted. The strongest wind gust I received at my house was 32 miles per hour. In Waukesha County, this same storm had produced 90 mile per hour gusts. The contrast shows us just how much a thunderstorm can dissipate when it is deprived of warm, moist air.
The conditions in the upper atmosphere are very vital to thunderstorm development. The day the derecho developed, conditions were in place that favored thunderstorm development in North Dakota. The storm activity rode the winds all the way down to southeast Wisconsin. To analyze upper air data, I reviewed upper air charts provided by the Storm Prediction Center, which are based on the atmospheric pressure in millibars. Each point on the upper air charts may not be at the same altitude.
For favorable thunderstorm development, the surface air chart should show high moisture levels and a source of lift. The dew points were in the sixties near Kenosha and Milwaukee around 7:00am. There wasn’t much change as the day progressed from 7:00am to 7:00pm. In the 925mb and 850mb maps, the same traits of high moisture levels and a source of lift should appear just as they do in the surface chart. The 850mb map shows moisture sitting to our south at 7:00am and also moisture extending up to North Dakota where the thunderstorms formed. At 7:00pm, moisture began pooling across Wisconsin in the 850mb and 925mb maps. The 700mb and 500mb maps indicated strong winds, which are vital to downburst speeds. At 7:00am, the winds were relatively strong across the Dakotas and Minnesota. At 7:00pm, there were very strong winds over Minnesota and western Wisconsin. In the 300mb and 250mb maps, there was a strong jet stream, which helped drive the storm in a southeasterly direction. Divergence, which is separating winds, is another thing to look for in the 300mb maps. There was strong divergence over the Dakotas at 7:00am, but it shifted to Iowa and southern Minnesota over the course of the day. Divergence works almost like a vacuum, helping to suck moisture in to initiate thunderstorm development and also helping to weaken the cap.
This derecho affected many communities across thirty-nine counties in Wisconsin. Mainly there was just wind damage, except for a few hail and tornado reports. Most derechos are mainly wind producers. Here are some of the most significant reports:
- Strong gusts in Waukesha and Washington Counties up to 90 mph.
- A building was moved by the wind at Highway 21 and Sand Pit Road west of Oshkosh.
- Tornadoes in Dodge and Buffalo Counties.
- Barron County had roofs blown off houses.
- Two and a half inch hail was reported in La Crosse about the size of tennis balls.
- Tornado Summary
At 9:13pm, a tornado touched down in extreme northwestern Dodge County in the township of Fox Lake, about four miles north of the lake Fox Lake. The damage started on North Parish Road, which is less than one mile away from the border with Green Lake County. It moved southeast near the state correctional institute at 9:15pm. It then moved just to the east of Fox Lake and the Fox Lake community, crossing Alto Creek and county road FF at 9:20pm. It crossed several more rural roads and died at 9:35pm at Buckhorn Road and Highway C about 7 miles west of Horicon Marsh.
The tornado was about 75 yards wide and it ripped through eight miles of rural area for about 22 minutes. The tornado had maximum wind speeds of approximately 100-110 mph. This made it about an F1 tornado, which is considered weak. This tornado mainly ripped down trees, crops, and small buildings like sheds. If it had occurred over a town, the results would have been different.