United States Army Corps of Engineers

Rock Island, Illinois

Our visit to the United States Corps of Engineers office in Rockport, Ill provided us with an understanding of the Corps' position on flood control and what they did during the 1993 Mississippi flood crisis.

Our visit involved a slide show presentation by Gary Loss (Corps engineer), tour of the GIS facilities, and tour of Lock and Dam 15. The slide show is presented here with slides and text paraphrasing the key points. Questions and Answers are denoted with Q and A.


The Rock Island District controls the river between Lock and Dam 10 to Lock and Dam 22. The Corps' purpose is to monitor the river changes, control flow, handle emergencies on river, and during off times give talks to communities and local governments on flood control and river management.

There was record flooding on the Mississippi River, Cedar River, Des Moines River and Iowa River all at same time. These coincident events are a "never-never" situation by Corps' standards. As engineers they decided from this event they need to have a broader scope on how to look at possible events like this in the future. To present, we only have the last 80 years on record as good river data, but we look at tree rings and floodplain deposits for other clues to a river's history.


Cause of flood: A weather pattern that did not move for two months brought the gulf stream into central midwest making Iowa the bullseye for activity. Cold air coming down from Canada caused much precipitation. What the Corps usually designs for is the record event on one river basin. The problem here was that there were several record setting events at one time spanning many river basins.


This slide shows the stage history since 1800 at Keokuk, Iowa. Note how far off from normal the Mississippi River was during the 1993 flood.

peak discharges to note:
1851: 316,000 cms
1973: 300,000 cms - considered the 100yr flood
1993: 470,000 cms


Mississippi River stage at Hannibal, Missouri for July 1993. Graph shows the rising and dropping stages with corresponding levee district breaks.
As each levee district breached the river would flood the district and the whole river would see a slight decline in stage, but within a few days the river was back up as if a levee district had not broken. People think that if that levee land was available the river would not have flooded as much. But the volume of water is so much, that it only takes a few days to fill a levee district then the goes river back to high levels again after district is full.
But... the breakages in the Fabius and the Sny levee districts (largest in the world) helped keep the crest down and possibly reduce the stage by several feet saving other levee systems.


The Corps had an emergency headquarters in Rock Island, Burlington, Quincy, and Des Moines running 24 hours a day seven days a week. They were responsible for dispatching info, answering questions, and handing out sand bags. We gave away 13 million sand bags in the Rock Island district during 1993, a normal flood event would have been about 1.5 million. The task we had was monitoring one hundred miles of levee that were three to four feet too low and figure out how to raise them. All the levees along the Mississippi are sand levees.


Levees were flooded from the back side as well as front. The pump stations could not keep up with incoming waters from local drainages and conditions for work were terrible. Slide shows levee holding back the Mississippi River to the right with boards and sand bags.


We used anything we could and received help from around the nation. Slides show a rented bulldozer and a Mennonite hauling hay on a ATV. One of the greatest sources of man power were the local prisons, using inmates to help save the levees.


One big task was tracking and controlling boils that come from underneath the levee. Pressure from river side of levee causes water to pipe under and up the land side of the levee. To stop it, we ring it with sandbags to control the runnoff and create a head to put pressure back on it. We watch for what it is pumping: if it is pumping clean, then we don't worry because it is just releaving pressure... if starts to pump material then we worry because it is eroding the levee from below - so we raise the ring to create hydraulic head and slow influx of water.
Pumps were brought on site to keep land side waters pumped into the river so the backside of the levee did not flood as well.

There was one levee break started by a man, who was later found and convicted. He broke the levee across from Quincy, Iowa on the Fabius levee. He said he did it because he felt the authorities were not paying enough attention to the levee so he moved a few sand bags and WHOOSH! Once it starts the sand just washes out. 400 homes were flooded and the access to Quincy was cut.


A gas station at Quincy was flooded both in 1993 and 1973. A few days after it flooded in 1993, a barge floated into the area and hit the gas station causing it to explode.

Many small towns were flooded.

In some areas the river was seven to eight miles across.

In Des Moines along the Racoon River, the discharge was 67,000 cms at 03:00am on July 11. The previous record was 42,000 cms. Several episodes of rainfall in different parts of the Racoon drainage overwhelmed the river's ability to carry waters and the water treatment plant flooded. 250,000 persons were left with out water. One of Corps' responsibilities is to get water out to the citizens via tankers and large jugs from local wells.


Sailorville Dam
Water flows over the spillway with a frequency of every twenty-five years. It washed out roads and carried silt and sands down river. The banks are on a 1:8 slope. This river joins the Des Moines river downstream. Part of the hillside is unstable and slides into the river when the embankment is eroded. Atop the hill is a town, which the Corps bought out at $250,000 per home.
Although Quincy and Keokuk were experiencing the highest discharge ever, all the attention was on Des Moines. At the Sailorville dam, surface waters were causing some inflation of clays. The public's perception was that dam was failing. The Corps covered the dam with plastic to control surface water infiltration but the public did not believe this would help. So the State of Illinois hired a private firm to check Corps' work.


Both the University of Iowa and Iowa City were both within inches of being overtopped and loosing water supply to thousands. This was occuring just days after Des Moines. The Corps spent the next month carefully throttling water through the dam and kept water within several inches from overtopping the levees. All the creeks feeding Iowa River were in flood stage too.


The Sny Levee District is the world's second largest drainage district. It is made from an old Mississippi River channel that was cut off. It protects very rich farm land. The Sny is divided into districts and we lost the upper portion of the Sny 25-30,000 acres.


This slide shows the districts we lost (green hatchured) areas saved (yellow solid). The lower Sny was saved, and a portion around Quincy was saved. The Quincy levee was saved primarily because it was raised from a 50 to a 200 year levee because of the intense industrial and commercial setup that the levee was protecting.


Normally levees are built on a 1:3 slope, built of clay, impervious to keep the water from coming through. The sand levees are very unique and we find them to be very practical but most Corps districts will not use them. We dredge the river sands and build levees with them. 1:5 slope on the land side and 1:4 slope on the river side so it is a relatively flat structure. Water will seep through levee releaving pressure. The trick in 1993 was to steal some sand from the base of the levee to add to the top of the levee without causing failure. This was successful and levees were lost only because the water overtopped the levee.

What did the CORPS learn?

That there were some areas that we did not know were as vunerable as they were, like the water treatment plants below reseviors.

We could not release as much water as we thought we could and that we need to work with the local governments more closely to decide when the waters should be released.

We are conducting a flood plain study in which we are trying to look at policy issues - are there places were we need projects to construct or raise levees?

Some places were we could not econimically build levees (like protected wet lands, small communities). So should we look at relocating the structures at risk. It is not worth $3-4 million to protect 320 people's residence, schools and business. Move them to higher ground insteasd.

We base 100, 200, and 500 year flood from statistics on data collected over the last 80-100 years, and maybe we should rethink how come up with these numbers. Right now we simply extrapolate a curve from discharge and stage measurements from historical data.

Even if we built all the levees half way back to the bluffs, there would still be railroads, highway embankments, and bridge constructions that will control the height of the river just like the levees did. So we have to think about all the river towns and having to move them back too... it is a major question for society - where do you want to have people? People want to be next to the river.