Founded in 1862, Union Pacific Railroad is currently one of the largest private landholders in the United States. Most of these properties support the ongoing operation of 32,000 miles of track in 23 states across the western United States. Union Pacific owns and manages additional property associated with discontinued rights-of-way and support facilities. Through the acquisition of other railroads, Union Pacific's present-day property ownership dates from the mid-1800's.

Valuation map along rail line in Colorado.

Managing all this property over the past 136 years has caused issues for the railroad. Valuation maps were drafted and maintained to help control the purchase and sale of property over the years, but numerous updates and the ravages of time have made these maps fragile and difficult to work with. Since the information contained on these maps must be used for making key financial decisions for the railroad, it is necessary to disseminate this information widely throughout the Real Estate Department and the company as a whole. These issues are compounded by the fact that there are 32,000 real property valuation maps in the railroad archives.

Eventually, the railroad came to the conclusion that it needed an effective way of handling these maps. It was the vision of John Hawkins, a project manager of Information Systems in the Real Estate Department, that GIS was the most viable solution to the problem.

"We came to the conclusion," says Hawkins, "that the best way to solve many of our problems was to digitize and spatially locate our maps. This way we could also relate newer, spatially enabled data with our historic maps."

The railroad basemap data is a highly accurate source created using Union Pacific's Precision Measuring Vehicle (PMV). PMV is a truck or sport utility vehicle equipped with cameras, lasers, and GPS units that can drive directly on the tracks and collect a very precise GIS representation of the rail bed. This was the data with which the maps were to be located.

Union Pacific Precision Measuring Vehicle alongside freight engines.

The railroad submitted a request for proposal that went out to competitive bid to spatially locate 15,000 of the valuation maps. These 15,000 represented the maps that fall along existing railways, which, therefore, would be the easiest to locate. Esri's Database Services Department, a group within Esri's Professional Services Division, was awarded the contract based on price; a sound solution presented at an on-site presentation at Union Pacific headquarters in Omaha, Nebraska; and Esri's reputation as a leader in the GIS industry.

The scope of the project included spatially locating scanned valuation maps of rail lines within 26 states of the western United States. Index polygons were then created around each georeferenced map to provide ease of access. The maps were aligned and indexed in ArcView using the georeferencing and editing tools. GIS Data ReViewer was used to perform quality control.

Managing all of these scanned images was challenging on many fronts. First, the maps themselves had names that were not necessarily meaningful to many of the people using them. The original name, which was based on state, railroad, and valuation section number, had to be replaced with a much simpler naming convention. The railroad elected to use a state abbreviation and an index number to support this. A link still had to be maintained to the original naming convention, which is how the engineers who update the maps keep them organized. The index polygons maintained both features, and the images were stored in a directory structure that preserved the state, railroad, and valuation section convention.

Index of georeferenced valuation maps along rail lines in the western United States.

Another challenge was keeping delivery data together while ensuring that no image was lost. The images were delivered to Esri on 30 CDs, then divided by state and delivered to the subcontractors that were to perform the work. The challenge arose when some images passed quality control while others did not. It was necessary to send rejected images back to the subcontractor for repair, but the tight schedule would not allow rework to hold up deliveries. The solution was to divide the data deliveries into groups for the subcontractors, quality control check the images, deconstruct the deliveries, and recompile them as new deliveries containing only accepted images for delivery to Union Pacific.

Verifying the quality of 15,000 images required a creative use of GIS Data ReViewer (an Esri Production Line Tool Set for ArcGIS component). The way GIS Data ReViewer is traditionally used, the quality control technician reviews a selected sample of features within a new database and identifies features that have errors. The errors are then recorded as new records in an error table. This approach is very effective for most quality control implementations but was not appropriate for performing 100 percent quality control on 15,000 images. Instead, Esri converted the image inventory for each delivery into a GIS Data ReViewer table. When an image was georeferenced, the technician would place a shape point at or near the center of the georeferenced image. The shape point was given the same object ID as the record in the GIS Data ReViewer table. Then, when the quality control technicians wished to locate an image for verification, they needed only to click on the record in the table and ArcView software's ArcMap application would "drive" the display to the correct location automatically.

Once the project was completed, the railroad began seeing immediate benefits. The Real Estate Department already uses the maps to geolocate encumbrances on its property, establish rights-of-way, and identify excess properties that can be sold off, among other things. According to Hawkins, the georeferenced maps "have saved an immeasurable amount of time doing property research."

"There are many more uses," says Hawkins. "All of the functions we do in the Real Estate Department depend on these maps. We have found even more uses for them than we had originally envisioned. Many entities within the railroad hierarchy are also finding great uses for the newer structure that these maps are in." One group is using the maps in ArcGIS to identify possible locations to build cellular transmission towers. Another group is using the maps to manage properties that can be leased for the installation of new communications trunks that run along Union Pacific corridors.

For more information, contact John Hawkins, Union Pacific (e-mail: jlhawkin@up.com), or Adrien Litton, Esri's Database Services Department (e-mail: alitton@esri.com).