Separating combined storm and sanitary sewer systems is something many cities eventually have to tackle. Lincoln, IL, faced the situation in a 13-block area on the city's west side, a fully developed area in the older part of town. The area was served by a combined sanitary and storm sewer system whose capacity was often overloaded during heavy storms, causing flooding on the streets and in basements.

A study of Lincoln's sewer system conducted in the early 1990s identified the area as problematic and proposed separating the sanitary and storm sewer systems. That meant disconnecting the stormwater inlets to the existing sanitary sewer system and diverting them into a new storm sewer system that the city would construct in the same general 13-block area. The size and complexity of the project made it one of the largest sewer construction projects that Lincoln undertook in many years, and some members of the city council were initially hesitant to approve it.

The city hired Farnsworth & Wylie Inc. of Peoria, IL, to study the situation and design the new storm sewer system. Designers ran multiple hydraulic analyses, evaluating different sewer-pipe locations to minimize material and construction expenses. They also used the computer model to optimize the new system within the constraints of an existing storm sewer. "I can't imagine solving this problem by hand," says Jeff Gastel, a project manager at Farnsworth & Wylie. "Yet with the computer model, we could design and optimize a complicated system that combined both old and new components." Extensive modeling to explore all the options saved an estimated 10% on the overall cost of the new system.

Gathering Data and Building the Model

One of the first tasks was to gather information to better understand the problem area and its existing sewer system. The city provided an AutoCAD plot of the infrastructure, which Farnsworth & Wylie supplemented with additional topographical data to confirm elevations of inverts of the existing storm sewer and sanitary sewer pipes. Gastel also walked the entire watershed area to determine the location of sub-basins and to get a feel for the density of the development and amount of impervious surface.

Although the earlier study of Lincoln's system had employed the Rational Method for calculating peak flows, the city specified the use of Technical Release 55 (TR-55) methodologies for this project. TR-55 presents simplified procedures to calculate storm runoff volume, peak rate of discharge, hydrographs, and storage volumes required for floodwater reservoirs. These procedures are applicable in small watersheds, especially urbanized watersheds. First issued by the Soil Conservation Service (SCS) in January 1975, TR-55 incorporates current SCS procedures.

Farnsworth & Wylie decided to expand its hydraulic software library and purchase a program capable of performing TR-55 analysis and complex routing methods, so another preliminary task was to find and purchase such a program. Gastel evaluated five programs and concluded that Hydra, from Pizer Inc. in Seattle, WA, best met the project's needs. The program models the hydraulics in municipal storm, sanitary, and combined sewer systems and gives the user the option of picking TR-55 or other routing methodologies. It can also make recommendations to correct for surcharges and problem areas in existing systems.

As much as possible, the city wanted to keep costs down by incorporating elements of the existing storm sewer into the new system. "We needed to do an analysis of the existing system to determine the capacity to which we could use it," Gastel explains. "We knew there were constraining segments in each sub-basin area. We needed to find these before we located the new system so we would know where additional capacity was needed." The computer model has the ability to make recommendations to correct for surcharges and problem areas in existing systems.

Gastel created a digital model of the existing storm sewer system by importing the city's AutoCAD file into the new modeling program. The AutoCAD file showed the network of pipes, and Gastel added information he collected earlier, such as elevations of the inverts and locations of manholes. Then he loaded the model with rainfall amounts generated by the 10-year, 24-hour storm as determined by the US Geological Survey. The software determined the flows from each of the sub-basins and graphically portrayed the movement of water through the existing system using hydrographs, which are typically bell-shaped plots showing the peak in-flow that occurs between hours 10 and 17 in this type of storm. The hydrographs clearly showed the points where the existing system was overburdened.

Fine-Tuning the System

With this information, Gastel began the horizontal layout out of the new system. This task was complicated by the fact that the route had to miss existing gas mains, fiber-optic lines, water mains, and so on. Initially, Gastel drew the system on paper, but once he had a good idea of the general horizontal configuration, he created a digital model of it in Hydra. He loaded this model with the same 10-year, 24-hour storm data used previously, then directed the software to calculate the inverts and the pipe sizes needed to handle this storm given the horizontal layout he had created. The results gave him all the parameters for a storm sewer system that would handle the required rainfall amounts in conjunction with the existing system.

Because it was easy to modify the computer model and repeat the analysis, he began fine-tuning the design with the intention of lowering the cost. For example, if he had several possible routes that a particular pipe could take to miss existing infrastructure, he evaluated them all to see which would cost less in terms of materials and construction requirements.

Gastel performed the analyses using the TR-55 methodologies as requested by the city. As a check on the system, however, he also ran an analysis using the Rational Method. "This gave us a lot of confidence in the results we were getting," he notes. "The Rational Method results compared as expected to the TR-55 analyses and provided a conservative check on the final design."

The numeric results of the analyses provided the quantitative support the city council needed to approve the Lincoln Sewer Separation project, as it was officially termed. The project was completed at the end of 1999, and by all indications the new system is working well. "This was our biggest sewer project in years," says Steve Mesner, the city's finance chairman when the project was in the planning stages. "Many people wanted assurance that the city would get a good return on its investment. The computer modeling was helpful for that, especially all the iterations that Farnsworth & Wylie did to get the best cost. The system has been built, and it works. We have not had the flooding or backing up that we used to."