This is the third in a series of three posts about the worst tornadoes to strike the Chicago-area in modern times. Two of the three happened on the same day in 1967, with one being just a few miles from where my family lived at the time (I was not yet born).
There wasn’t a tornado warning… there wasn’t even a tornado watch. On August 28, 1990 an F5 tornado struck Plainfield, Illinois killing 29 people and injuring 350. 470 homes were destroyed and 1000 more were damaged. Damages totaled an estimated $160 million. It is the only F5 tornado recorded in August and the only F5 tornado to strike the Chicago area. There are no known videos or photos of the heavily rain-wrapped tornado.
At 10am on August 28, the National Severe Storms Forecast Center in Kansas City (now the Storm Prediction Center in Oklahoma) upgraded their severe storm outlook for northern Illinois from a slight risk, issued earlier that morning, to a moderate risk. At 1:28pm the NSSFC issued a severe thunderstorm watch for portions of northern Illinois.
Conditions were favorable for severe thunderstorms. A cold front was forecast to move southeast across northern Illinois during the afternoon. Ahead of the cold front, the atmosphere was unstable with CAPE (Convective Available Potential Energy) values over 4000 J/Kg when thunderstorms initiated near the Wisconsin/Illinois border around noon. However, weak wind shear in the lower atmosphere suggested more of a wind/hail threat than for tornadoes.
The atmosphere became more unstable during the afternoon as temperatures soared into the 90’s and dew point temperatures reached the upper 70’s. Thunderstorms grew to a height of 65,000 feet and began to exhibit supercell characteristics. As the Plainfield storm moved southeast the low level wind shear increased. By 3pm, CAPE values had climbed to an astonishing 7000 J/Kg. A series of downbursts and funnel clouds occurred in rural southern Kane County before the increasing wind shear combined with the intense updraft eventually led to the development of the main tornado near Oswego in Kendall County.
The tornado reached its peak intensity between 3:15pm and 3:45pm as it moved from Kendall County into northwestern Will County and through the communities of Plainfield and Crest Hill. After the main tornado dissipated in Joliet, the parent thunderstorm continued producing damage as it moved through Kankakee County and into Indiana. The storm produced nearly continuous damage over northern Illinois for 4 1/2 hours.
I started the day like I had started the day so many other times that summer – in hopes of seeing a good storm. August 28, 1990 started out warm and very humid. A very unstable air mass promised big storms but the lack of vertical shear led everyone to believe that storms would be big and rainy but not tornadic.
After departing soon after the noon weather report by WGN’s Tom Skilling, I took off from my home in Glen Ellyn to an area just northwest of DeKalb. I observed a storm very similar to storms I had seen often during the summer of 1990. There were reports of golf ball hail to my north, but nothing in my immediate area indicated what was soon going to become of this tremendous storm. I worked my way down through DeKalb where rain and hail started to fall. I struggled through worsening conditions onto eastbound I-88 where I decided to head south on Route 47. As I pulled out of the rain near the Sugar Grove airport, I was amazed to see the storm change it’s characteristic of being a big rain producer to being something a bit more scary! It was a little after 3pm when I started video-taping a very intense-looking storm about to cross the road to my south. Torn between what I now know to be a rear flank shelf cloud and an area almost directly on top of me just to my northwest, I decided to try to head south to Route 126 and head to Plainfield. Unfortunately it was not to be that day. The rear flank of this now 65,000 foot tall storm surged across the road, knocking down power lines right behind me and doing a significant amount of damage at the airport. As the driving conditions became impassible I took shelter in a BP station on the corner of Rt. 47 and Rt. 30. People were all commenting about the storm. I had no cell phone at the time, and the phone in the station was being used. I remember thinking to myself “This storm might produce a tornado! I’ll bet that someone is reporting this!”
I was right with my first prediction but terribly wrong with the second. The storm soon put down the tornado just to the southeast of my location. It was not until a few hours later that I became aware of how significant of an event this was.
-Paul Sirvatka (video below)
The tornado formed from a supercell thunderstorm which initially formed near Janesville in south-central Wisconsin, producing a tornado near Pecatonica in Winnebago County, Illinois (near Rockford) which touched down at 1:42pm. Continuing to move to the southeast, the supercell spawned the Plainfield tornado which touched down at 3:15pm near Oswego in Kendall County. The tornado rapidly strengthened into a violent, rain-wrapped tornado, traveling southeast into Wheatland Township of Will County, near the Wheatland Plains subdivision northwest of Plainfield, damaging many homes and causing several injuries but no fatalities.
Past Wheatland Plains, the tornado continued to strengthen as it tore across open farmland, and reached F5 intensity in this area. A narrow swath of very intense ground scouring was observed, as mature corn crop was completely stripped from the ground, leaving nothing but bare soil behind. Several inches of topsoil were removed as well. As the tornado crossed US-30,a 20-ton tractor-trailer was thrown more than half a mile from the road, killing the driver. Three other motorists were also killed in this area as their vehicles were thrown from the road.
The tornado struck Plainfield around 3:28pm at high-end F4 intensity. Around 3:30pm the tornado directly struck the Plainfield High School, killing three people, including a science teacher and two maintenance workers. Students who had been out practicing for the fall football programs ran into the high school to take shelter a few minutes before the storm hit. After an alarm was pulled by a dean in the main office, the volleyball players preparing for a game in the gymnasium rushed to the nearest door and took shelter in the hallway. It is reported that as soon as the last player was through the door, a coach quickly closed it, only for it to be immediately ripped back off by the storm. They gymnasium proceeded to fall apart and crash down filling the gap in the doorway. They took shelter in the same hall as the football team, and once the tornado had passed, that was the only hallway left standing in the building.
The tornado then demolished the Plainfield School District Administration building, where the wife of a custodian was killed. The tornado crossed Route 59 (Division Street) and ripped into St. Mary Immaculate Church and school, claiming an additional three lives, including the principal of the school, a music teacher and the son of the cook at the rectory. 55 homes were destroyed in Plainfield itself, some of which were swept away. A grocery store east of the high school was badly damaged, gravestones in a nearby cemetery were toppled and a metal dumpster was found wrapped around the top of a partially debarked tree. Damage in Plainfield was rated as high-end F4.
The storm worked its way southeast towards the large city of Joliet, damaging homes in the Crystal Lawns, Lily Cache and Warwick subdivisions killing five more people. An additional three people would later succumb from injuries sustained during the tornado. The tornado ripped through the Grand Prairie Elementary School causing significant damage. Observers watched from the doors at the Louis Joliet Mall as the tornado passed just southwest of them. Many more homes were destroyed in Crystal Lawns, Peerless Estates, Lily Cache and Warwick – many of which were newly built.
The tornado then moved towards Crest Hill. At 3:38pm the storm ripped through the Crest Hill Lakes Apartment complex, causing F3 damage and claiming another eight lives, destroying one apartment building and half of another. Neither have been rebuilt. The tornado also ripped through the Colony West subdivision, destroying town homes, and killing a married couple in their car on Cedarwood Drive outside the apartment complex. The tornado then destroyed three apartment buildings on Elizabeth Court and more homes in Bridalwreath, southeast of Elizabeth Court. Homes were damaged on Arden Place, and two high tension wire structures were destroyed at Black Road and Arden/Mason. They were replaced with different structures than the previous two and can be seen as such today.
Further southeast, the tornado lost strength until lifting near Woodlawn Avenue and Campbell Street in Joliet. The parent thunderstorm continued until it crossed over the Indiana border, dissipating around 4:30pm.
In the months following the tornado, the National Weather Service was heavily criticized for providing no warning of the approaching tornado. The NOAA Disaster Survey Report was highly critical of the forecast process within the Chicago office as well as coordination with local spotter networks and the preparedness of these groups. Prior to 1990, the National Weather Service in Chicago was responsible for providing forecasts for the entire state of Illinois.
As the Chicago office was overwhelmed with its workload, no warnings were issued by the office until 2:32pm – nearly an hour after the first tornado was sighted southeast of Rockford. A second severe thunderstorm warning was issued almost an hour later at 3:23pm, but this provided no indication that a tornado was on the ground and did not mention the area where the tornado had tracked. No tornado warning was issued until after the tornado lifted.
According to the Chicago Tribune, NWS Chicago had “the worst record in the nation” for issuing warnings for severe storms, only issuing warnings 24% of the time when a warning was justifiable. Even today, many meteorologists refer to the “Plainfield Syndrome” as the idea that it’s better to issue too many warnings and be wrong than to miss one critical warning, as was the case for the Plainfield tornado. At some point after 1990, the National Weather Service reduced the Chicago office’s workload by creating an office in Romeoville, Illinois and allowing offices in the Quad Cities, St. Louis, Indianapolis and Paducah to issue forecasts for their respective areas.
The tornado that devastated portions of the southwestern suburbs of Chicago on August 28, 1990 came as a surprise to forecasters and the general public alike. During the past 20 years there have been so many incredibly important advances in the science of meteorology, and in particular in forecasting severe weather and tornadoes. Some of the data that is available today would likely not have been even dreamed of by many forecasters 20 years ago. In addition to a greater understanding of how and why tornadoes form, meteorologists today also have a wealth of observational data that was not available back in 1990 as well as far more accurate and higher resolution numerical models.
One of the newer computer models that is getting a lot of use by both operational forecasters and researchers is the WRF model (Weather Research & Forecasting model). One of the nice features of the WRF model is that it can easily be used for research, allowing meteorologists to run the model for past events. The National Weather Service office in Chicago modeled the Plainfield event to see if the WRF would do a better job of identifying a greater tornado potential.
The WRF model forecast generated using data three days prior to the tornado really did not suggest any appreciable tornado threat. In fact, the forecast showed a cold front down to the I-70 corridor by afternoon and no thunderstorms developing within 300 miles of Chicago. While the atmosphere was forecast to be hot and very unstable, the model was maintaining a strong cap (layer of warm air aloft preventing thunderstorms from forming).
The WRF model’s forecast for the event using data the day before the tornado was an improvement. The front was forecast to move slower and still be over northern Illinois in late afternoon, and the WRF was also showing a much weaker cap and thunderstorms developing over northeast Illinois and Indiana, Ohio and southern Michigan. Still, the parameters did not suggest much of a tornado threat, let alone conditions for an F5 tornado to occur.
A subsequent run of the model using data from the morning of August 28 really did not show any appreciable change in the forecast. The forecast data produced by re-running of the WRF did improve somewhat as the event drew nearer, but there was never any indication using the WRF model that there would be an enhanced tornado threat for northern Illinois. While this is not the result that was hoped for, all hope is not lost. The forecasting of tornado potential does not rely solely on forecast model data. As an event nears the forecast hinges much more on observational data such as satellite, radar, upper air soundings and surface observations. The other thing to consider is that the WRF model was run using data that was available in 1990, and that 20 years later much more observational data is fed into the WRF model, which would likely result in a different and possibly more accurate forecast.
*** Sources include NWS Chicago “The Plainfield Tornado of August 28, 1990”, Paul Sirvatka, Jay Orbik (Youtube videos), Wikipedia and Gino Izzi’s “Examining the Forecastability of the Deadly Plainfield F5 Tornado Using the WRF Model”.