A record setting 24-hour rainstorm hit Chicago in September 2008. It created an opportunity to assess its impacts and to compare the storm water management with that used in other recent rainstorms. Flooding impacted all forms of transportation and thousands of homes and businesses, 10,000 homes had to be evacuated. One lesson learned is that this storm's magnitude was enhanced by urban and lake influences on the atmosphere that extended over the large suburban areas west of the city. This helped intensify the rainfall over a large area. A second lesson learned is when 6-hour and 24-hour rainfall amounts exceed once-in-25-years recurrence values, the water management facilities cannot handle the large volume of water. Hence, major diversions of flood waters were needed to be made into Lake Michigan and the Illinois River.
Chicago and its suburbs experienced a record heavy 24-hour rainstorm on September 13, 2008. Rainfall reaching 16.9 cm, exceeding recurrence interval values of once in 50 years, occurred over 485 square kilometers. This flash flood quickly created massive flooding that stopped surface transportation, caused evacuation of 10,000 homes, and led to property damages totaling $155 million. The city was aware of the approaching storm and acted to remove waters stored in the city's deep tunnel storm water reservoir system. The emptied tunnel waters plus early stormwater totaling 49 Mm3 were released into the Illinois River. However, the management system could not handle the extreme volume of storm water, and as urban flooding developed, managers chose to release 341 Mm3 into Lake Michigan, creating water supply problems from the polluted storm water. The impacts from the urban flooding included all forms of transportation, thousands of flooded homes and businesses, and 10,000 homes had to be evacuated. This storm and the management actions employed illustrate the sizable difficulties large cities face in handling waters from major rainstorms.
Chicago's official raingage at O'Hare Airport recorded 16.9 cm (6.64 inches), a new Chicago record for a 24-hour period. The atmospheric remnants of Hurricane Ike moved northeast and crossed central Illinois on September 12; the passage of its low pressure center with abundant moist air along a cold front boundary resulted in heavy rainfall across northern Illinois (Figure 1). Amounts ranged from 5 to 7 cm eastward from the Mississippi River. The rains intensified on September 13 in the Chicago area, with some amounts exceeding 15 cm.
This record-setting rainstorm at Chicago presented an opportunity to examine two key questions. First, what were the types and magnitudes of impacts of the event? Second, what water management actions were used to handle a record amount of water in the urban area? The record-setting rainstorm in Chicago in September 2008 also provided an opportunity to examine the effects of improvements since 2001 in Chicago's water management and treatment system. Flood waters from two severe rainstorms in 2001 (Changnon and Westcott 2002a) overwhelmed the city's developing deep tunnel-reservoir storage system, but the system has been enlarged since 2001. The city had also developed a Rain Blocker system, at a cost of $75 million, to slow the inflow of flood waters into the sewage treatment system, prolonging time for water treatment to occur. This system and the deep tunnel system had not experienced extremely large rainfall amounts since the two 2001 storms. Chicago has long had serious water problems, some related to the amount of water diverted from Lake Michigan, and some related to the management of storm runoff.
Major previous rainstorm impacts have included extensive flooding of residences and businesses, and major problems for transportation systems in the city. The vast quantities of storm water in September 2008 led to the flooding of several major highways, many streets, and rail lines. The storm waters also flooded thousands of buildings, leading to evacuations of more than 10,000 homes. Property insurance losses totaled $155 million, identified as a catastrophe by the insurance industry (Property Claim Services 2008). These losses included damages to homes, businesses, and vehicles. Stormwater storage was inadequate to handle the vast quantity of water, and diversions of stormwaters away from the city were necessary to lessen the flooding.
Rainfall totals in Cook County, home of Chicago, for the storm period that began between 1900–2100 CST on September 12 and ended by 2200 CST on September 13, are shown on Figure 2. Rain began first in the southern urban area (1900 CST) and ended latest in the north sector of Cook County at 2200. This rainfall pattern is based mainly on a network of 25 recording raingages (Westcott 2008). The 24-hour rain pattern is also defined by data from the National Weather Service (NWS) raingages in Cook County (at O'Hare Airport, Midway Airport, and Chicago Heights) and those in the surrounding area, mainly suburbs. The 25-gage network consists of recording raingages spaced in a grid with 10 km between gages, and these allowed assessment of peak rainfall amounts for periods during the storm. The average storm rainfall over Chicago was 11.3 cm, and the average over Cook County was 12.6 cm.
The record setting rainstorm on the 13th was preceded by light rains that occurred in midday on the 12th. Amounts across the Cook County area on the 12th were 0.25 to 0.63 cm. Rain began again during the morning of September 14 as a cold front passed through the area. This system produced 2.5 cm of rain in northern Cook County, and up to 6.4 cm in the south end of Cook County. This rain added to the flooding problems created on the 13th. The rainstorm on the 13th was the third rainstorm during 2008, five less than the record of eight rainstorms set in 2001 (Changnon and Westcott, 2002b). A moderately heavy rainstorm (5 to 7 cm) occurred in the Chicago area in August 2007 (Angel and Changnon 2008).
The heaviest rainfall on the 13th occurred in the northern part of the Chicago area where amounts ranged from 15 cm up to 18.2 cm at gage #3 (Figure 2). Amounts in the central section of Cook County were less, 7.5 to 10 cm, and amounts were heavier in the southern sections of the urban area where totals exceeded 12 cm.
This high-low-high sequence was also found in the patterns for the maximum 6-hour and 3-hour rainfall amounts. The 6-hour maximum amounts exceeded 10 cm in the northern sector and were 7.5 cm or more in the south and west, with 5 cm in the central portions of the city. The peak 6-hour rainfall occurred from midnight on September 12 to 0600 local standard time on September 13. The peak 3-hour amounts had values greater than 5 cm in the northern and southern parts of the urban area with 2.5 cm in the central section.
The statistical frequencies of the maximum rainfall amounts were determined based on recently developed values for the area (Huff and Angel 1992). The 24-hour rainfall amounts revealed that the 18.2 cm at gage #3 rated as a once in 100-year event, and amounts at six surrounding gages were rated as 50- to 75-year events. In the city's center area, the amounts rated as once in 5-year occurrences, with once in 25-year amounts at five gages in the south. Peak amounts for 12-hours had similar recurrence interval values. The maximum 6-hour amounts achieved recurrence intervals of 25-years in the north, whereas those in the center of Chicago had 5-year recurrence interval levels.
The temporal distribution of rainfall during the storm revealed that the heaviest amounts at all gages occurred during the 2400–0700 LST period. Such nocturnal maximums of rainfall occurred in 87 percent of all flash floods in Illinois (Changnon 1978). This timing meant that when many residents awakened, they were surprised to find serious flooding in and around their homes. Many commuters unexpectedly ran into flooded streets and underpasses, leading to traffic jams, accidents, and great delays. The remaining rainfall occurred between 0600 and 2100 on September 13.
Urban and lake effects on the atmosphere at Chicago often act to increase precipitation (Huff 1995). A 3-year meteorological field study of weather conditions in and around Chicago identified how urban effects, sometimes aided by lake effects on the atmosphere, acted to increase warm season rainfall and often led to heavier rains in heavy rainstorms (Changnon 1980).
An issue relevant to the 2008 storm relates to climate shifts that lead to more rainstorms over time. Studies of historic floods and flood-producing rain events in the Midwest revealed a temporal increase since 1930 (Kunkel et al. 1993). Assessment of heavy rainstorms in the Midwest since 1930 found increasing frequencies. The three 2008 storms, plus the eight in 2001 and one in 2007 at Chicago, reflect this climate-related increase in storm frequency, an important issue for water management planning and operations. Studies of global warming and its effects on climate concluded that heavier rainfalls would continue to increase in the 21st century (National Assessment Synthesis Team 2000).
Some past studies addressed the variety of problems that heavy rainstorms caused in Chicago (Changnon 1980). A common problem relates to flooding of viaducts and streets. Chicago is the national hub for commercial air traffic, the nation's railroads and trucking industry. Hence, heavy rains that disrupt and slow or stop traffic flow create large costs. Flood waters also disrupt movement of the city's commuter rail and bus systems, another major problem. Floods also have always produced extensive damage to property and businesses, and are a major challenge for the water treatment facilities.
The heavy September 2008 rains, which largely fell in a 6-hour period early on the 13th, created flash flood conditions and major impacts resulted. Record high river flows and allied flooding quickly resulted from the heavy rains, particularly along the rivers in the urban-suburban area (Figure 2). The flooding was particularly bad along the North Branch of the Chicago River, and also along the Des Plaines, Du Page, and Fox Rivers (Figure 2). The Illinois River at Morris (Figure 2) reached 7.55 m on the 16th, exceeding the flood stage of 4.9 m, and breaking the old record of 7.1 m set in 1957. This created flooding many kilometers from the Chicago metropolitan area. The Illinois River at La Salle (Figure 1) peaked at 7.1 m on the 17th (1 day later) which was higher than its flood stage of 6.1 m. The Des Plaines River (Figure 2) crested at 3.1 m at Riverside, a suburb, much above the flood stage of 0.9 m, creating flooding along the river in the western suburbs. The Fox and Du Page Rivers also flooded along their courses, creating damages along the Fox from Elgin (Figure 2) southward to Yorkville and Ottawa. Flooding was bad along the Du Page at Addison and other suburbs (Figure 2). The mean flow of the lower Fox River was seven times greater than the average flow. Similar flooding in the western suburbs also occurred in August 2007 after several days of moderate to heavy rains west and north of Chicago (Angel and Changnon 2008).
Flooded homes, with extensive damages in basements, caused more than 10,000 homes to be evacuated and thousands of persons were placed in shelters or homes of friends. Many evacuees were housed in the four Red Cross shelters in Chicago. Insured property losses to 22,500 homes in the flooded zones totaled $125 million. Communities with sizable evacuations along the Des Plaines River (Figure 2) included Glenview (3,000 persons), Des Plaines (1,000), Rosemont (250), and Riverside (400). Along the North Branch of the Chicago River, 2,800 persons had to be evacuated. Evacuations also occurred at Sugar Grove (75 persons), Morris (144), and Addison (750). Cleanup of flooded homes and damaged areas was done largely by local communities, and these efforts took three months (Chicago Tribune October 21). Cleanups cost $3.5 million in Cook County, $1.4 million in Will County, and $547,690 in Kendall County.
Several major transportation arteries were blocked. Two major commuter rail lines were closed for 3 days, and several railroads had freight trains delayed for 2–3 days (Trains 2008). Numerous streets, highways, and Interstates were closed for several days. Interstate 80–94, one of the region's busiest, was blocked for 4 days, and I-94 was closed in north Chicago (Chicago Tribune September 17). Closed streets led to major traffic congestion on non-flooded streets and highways, and thousands were late to work. Flood damages to vehicles created 5,500 insurance claims and amounted to losses of $18 million. Several city and suburban schools were closed from September 13th to the 16th. Commercial airline flights at O'Hare and Midway Airports were delayed 6 to 8 hours on September 13.
The Governor of Illinois acted after the flooding began and identified seven counties as Disaster Areas. These included Cook, DeKalb, DuPage, Grundy, Kane, La Salle, and Will (Chicago Tribune October 21). President Bush later identified the seven counties as Federal Disaster Areas, allowing residents with flood damages to seek federal aid. These seven counties were where the heaviest rain fell (Figure 2) and in downstream areas (Will, Grundy, and LaSalle counties) where massive flood waters had moved from the Chicago area. In mid-November the federal government began providing disaster aid to local government agencies in nine counties to cover 75 percent of the costs for the removal of flood debris, to repair streets and roads, and to repair public properties. Five weeks after the flood, thousands of persons were still relying on food stamps to cover losses from the flood. More than 5,000 appeals had been approved for federal disaster aid by October 20 (Chicago Tribune October 21).
Water management endeavors reflected reactions to the evolving flooding on the 13th. Chicago has constructed a deep tunnel and reservoir storage system, which consists of 109 miles of large tunnels under the city's rivers plus several surface reservoirs. Their construction cost $3 billion. Normally, storm waters stored in the tunnel system are extracted, treated at sewage plants, and then released into the Sanitary and Ship Canal which connects the Chicago River to the Illinois River.
Chicago has a peculiar storm water management situation related to the diversion of waters from Lake Michigan (Changnon and Changnon 1996), and the reversal of its in-lake drainage system. The city's three river exits into the lake are blocked by movable gates, and water taken in from the lake, and from rain falling over the city, is moved down the Illinois River. The annual amount of water diverted by Chicago for various purposes (local water supplies and flow down the Illinois River to dilute pollution and maintain barge traffic) is under federal control and cannot exceed a fixed level. Thus, to handle storm water, the city has three choices: 1) lower the waters in local rivers and the storage system in advance of a storm and divert it down the Illinois River; 2) store storm waters in its special deep tunnel system; or 3) divert flood waters into Lake Michigan. The in-lake diversion is undesirable because it moves polluted storm water into the lake where waters are withdrawn for city and suburban domestic supplies. The pollution of lake waters led water treatment plants in and around Chicago to add chlorine to minimize any problems with the potentially polluted water (Chicago Tribune September 16).
Research directed at storm forecasting and nowcasting of severe rainstorms at Chicago has shown that it is possible to forecast and thus anticipate rainstorms (their time, location, and amount) with enough time before the storm to lower the waters in the Chicago River system (Huff et al. 1980). This action was used to direct treated water down the Illinois River prior to the start of the September 2008 storm.
In response to the September 2008 rainstorm, the city filled the deep tunnel system to its capacity with 9.5 M m3 gallons of water, but that did not reduce the rapid increase in flooding. Cook and DuPage counties received rainfall totaling 469.4 M m3 of storm water. Chicago began releasing untreated flood waters into Lake Michigan on the morning of the 13th, and by the morning of the 15th had released 340.7 M m3 of flood water into the lake. An additional 49.2 M m3 of treated water were released down the Sanitary and Ship Canal to the Illinois River, done prior to the storm to help enhance in-city storage of the storm waters.
These 2008 diversions revealed that the recent improvements in the tunnel-reservoir system since the 2001 storms were not sufficient to handle the flood waters from a storm of the magnitude of that in September 2008. The 2008 storm had a flooding outcome similar to those with two extremely heavy rainstorms in 2001 (Changnon and Westcott 2002b). All three storms produced excessive flooding in the city and its suburbs. All three storms had 6- and 24-hour rainfall amounts that equaled or exceeded the 50-year return values and that covered more than 450 square kilometers. In contrast, six other rainstorms in 2001 with rainfall amounts at the 5- and 10-year return levels did not create any serious flooding. Thus, even with an extensive flood water storage system, Chicago's system can not handle the water generated by rains achieving the 50-year level over a large area, and the only option is to divert untreated water into Lake Michigan.
The record high 2008 rainstorm in Chicago produced the highest rainfall amounts in the northern metropolitan area, least in the city's center, and heavier rains in the city's southern sections. The resulting huge mass of water amounted to 469 M m3. The temporal distribution of rainfall during this 24-hour rainstorm began with light rain, followed by the heaviest rains for 6 hours at night, and then moderate amounts until the storm ended 12 hours later. Overall, the urban area averaged 12.6 cm of rain, and the stormwater management actions could not halt some flooding, particularly along the region's rivers. Storm impacts were similar to those in past heavy rainstorms with extensive home flooding and creation of major transportation problems. However, the number of evacuations was larger than ever before and the problems were in the city and the suburbs west of the city. Insured property losses amounted to $155 million.
The magnitude of the September 2008 storm and the use of diversions of flood waters into Lake Michigan and the Illinois River illustrate what will be needed when major rainstorms occur at other large cities. They should be prepared under heavy rainstorm situations to divert stormwaters into an adjacent lake, an oceanic body, and/or a large adjacent river system.
The September storm reveals the large and costly deep tunnel-reservoir system in Chicago is not adequate to handle the magnitude of water produced by rainstorms that create 6 to 24-hour amounts having once in 50-year or greater frequencies and extending over several hundred square kilometers. Lowering the river system by added inflow to the Illinois River before a rainstorm strikes the area will help but is not sufficient to handle waters generated by more major storms. Thus, future major rainstorms in the Chicago area will require sizable releases of water into Lake Michigan to minimize flooding in the city.
The property loss data were provided by Gary Kerney of the Property Claims Service, and Dan Injerd of the Illinois Natural Resources Department provided valuable information about the water management actions. Maria Peters and Nancy Westcott of the Midwest Regional Climate Center provided extensive rainfall data, and Eileen Deremiah assisted in the analysis.
Author Bios and Contact Information
Stanley A. Changnon has pursued and directed atmospheric and hydrospheric research for 57 years. He directed the atmospheric research and services program of the Illinois State Water Survey for 15 years and served as the Survey's Chief for six years. As Chief Emeritus, he also serves as a Professor of Geography at the University of Illinois. His interests include: hydrology; weather and climate extremes; climate variability and change; atmospheric effects on agriculture, water resources, and society; and accidental and planned weather modification. He can be contacted at the Illinois State Water Survey, 2204 Griffith Drive, Champaign, IL 61820. Telephone (217) 586–5691. Email: email@example.com.