ArcNews Online

Fall 2008

A Solution for Small-Scale Water Distribution Networks

DuPage Water Commission Implements Pipeline Data Model


  • The ArcGIS Pipeline Data Model stores features of pipelines.
  • The model helps account for linear referencing of features.
  • With the model, dynamic events and static features can be held.

A 12-foot-diameter tunnel and nearly 200 miles of pipeline carry water from Lake Michigan to communities in Illinois' DuPage County, just west of Chicago. The DuPage Water Commission, an independent unit of government created by special legislation in 1985, supplies municipalities, private utilities, and the Argonne National Lab. The commission's $400 million pumping station and water pipeline system fulfilled an idea dreamed up in the 1950s when residents in the burgeoning 300-square-mile service area noted a declining quality and supply of well water.

  click to enlarge
Custom ArcIMS Web site allows employees to search and query attribute data, print maps, and access specific as-built drawings based on the station values of the area selected.

The commission began operations in 1991. By 2004, the decision was made to upgrade the existing information management system of hand-drawn maps, road atlases, and "corporate memory." Following the recommendations of an engineering firm hired to do a needs assessment and implementation plan, the commission defined the GIS layers and datasets it needed, consolidated data, and hired GIS coordinator Frank J. Frelka.

With a background in public works, Frelka first attempted to adapt the system to one of Esri's water distribution data models, a collection of objects, feature classes, and attributes defined for water and wastewater distribution networks. Typically, the model can be tailored to meet project requirements for data management and integration with other systems. The DuPage water distribution network, however, was not typical.

"It certainly made sense to me to use a water distribution model, but it was not a good fit," Frelka says. "We had 27 customers instead of thousands. The feature classes, such as service laterals, fire hydrants, and commercial and residential customers, didn't apply to us."

Remembering a presentation at the Esri International User Conference a few years back, Frelka took a look at data models on the Esri Web site and discovered the ArcGIS Pipeline Data Model (APDM). APDM was designed for storing information pertaining to features found in gathering and transmission pipelines, particularly gas and liquid systems. After reading the white paper and studying the model, Frelka decided APDM was the solution DuPage needed. It met the commission's primary objective to account for linear referencing of features. From APDM's 50 feature classes, Frelka extracted the 15 he needed and defined an additional 20. Because water transmission runs under a less intense regulatory environment than flammable fluids, Frelka used a small fraction of the APDM attribute fields.

"With APDM it is possible to select and work with only the features you need," Frelka explains. "With the model, dynamic events and static features can also be held. I realized that events are used for dynamic situations. This was a static situation."

Starting with as-built drawings of the system, Frelka converted data into the GIS. He "eyeballed" pipelines using DuPage County centerlines and aerial photos. The pipelines were snapped to GPS points taken by survey every 500 feet and at every visible structure, such as manholes and electric cabinets. He created routes from the pipeline features; this involved a pipeline calibration using more than 400 known locations of valves and manholes.

"The lines were converted to routes, and we assigned measures to the lines," Frelka says. "Then we had to learn about calibrating routes working from known locations along the route with associated station values based on linear feet. There are several different types of stationing, but the most practical was flat 2D linear stationing. The surrounding terrain in the area is essentially flat. Things get complicated with station equations and routes."

Once Frelka had the routes, he developed event tables of asset locations from the as-built drawings—elevations, station values, valve locations, manholes, bends in the pipe, blow-off valves, system valves, CP test stations, and casings. He made the table in Microsoft Excel and imported it into ArcGIS Desktop. With linear referencing, Frelka was able to develop an ArcIMS application that opens as-built drawings for the specific section of the pipeline selected.

"I started with event tables, because all we had was tabular data," Frelka comments. "Using an event table gave us the ability to create feature classes. The system is static, which lends itself to out-of-the-box tools. Then I read the Esri Press book Linear Referencing in ArcGIS and went through the exercises. I found it was exactly what I needed for what I had to do."

More Information

For more information, contact Frank J. Frelka, DuPage Water Commission (e-mail:, Web:, or Peter C. G. Veenstra, director of product development, Eagle Information Mapping, Inc. (e-mail:, Web:

Contact Us | Privacy | Legal | Site Map