Simplifying Complexity with Scanning

Flood Protection Project in Cedar Rapids, Iowa a Good Test for Surveying Technology and Processes

By Mary Jo Wagner

With survey projects, there’s complex, there’s very complex, and then there’s the Cedar River Flood Control System (CRFCS) project in Cedar Rapids, Iowa. Launched in 2015, the 20-year, $750 million construction project is building a combination of permanent flood walls, removable walls, levees and gates along seven miles of the river’s banks.

As a survey consultant for the CRFCS, engineering company Foth Infrastructure Environment has been providing survey data for the project when needed. In 2017 it was tasked with an assignment that seemed innocent enough: survey one site to acquire railroad pier features for hydraulic modeling and analysis and perform an as-built, topographic survey at another site for flood wall design and clearance verification. Upon further inspection, however, the challenges to meet that assignment went beyond complex: Site One was only accessible by boat and involved an active railroad bridge and for Site Two, they would only have four hours to capture one mile of as-built data.

“For the setting, time constraints and required precision of the data, we had to have extremely efficient and reliable technology that could give us long-range data capture without sacrificing accuracy,” says Jody Budde, a professional land surveyor with Foth, based in Cedar Rapids. “Conventional survey techniques would challenge us to meet those specs and timelines and would put our crews’ safety at risk.”

Armed with a tactical fieldwork plan, GNSS and advanced scanning, imaging and high-speed scanning technology, a small crew not only captured all the infrastructure elements needed, they delivered more data than was initially requested and debunked a previous assumption about the true low point of a bridge arch, saving the city from a costly oversight.

Tackling Complexity

Planning for the CRFCS began while the last drops of water were receding from the “Flood of 2008,” the most devastating flood in the region’s history. In June of that year, the Cedar River crested at 31 feet, surpassing the previous record set in 1993 by 11 ft. The flood waters deluged 10 square miles – 14 percent of the city – displaced 10,000 residents and swamped Cedar Rapids, Iowa’s second-largest city, with $6 billion in damages. In 2015, the city adopted the CRFCS master plan and started redeveloping the riverfront and building the flood protection elements in earnest.   

Along with the levees, and permanent and removable flood walls, the CRFCS will add up to eight pump stations, replace one bridge, create a basin to store excess rain water and construct an amphitheater designed to be a storage basin during high water events. Taken together, the entire flood-protection system is designed to withstand the 2008 flood-volume levels. To date, seven phases of the long-term plan have been completed and four projects, including the first permanent flood protection control on the west side of the River, are underway.

Critical to designing the new flood structures was understanding the Cedar River bank topography and the existing infrastructure along the bank and in the water to accurately model hydraulic flows and determine volume tolerances through specific sections of the River. When Foth was tasked with acquiring this information, it became clear early on that they would need to move beyond conventional total station and scanning technology to deliver the precise detail both on spec and on time.

“In preparing for surveying both sites, we studied the areas of interest and the project specifications and we scanned one of the bridge arches with our existing equipment as a test,” says Budde. “Although we acquired the data, it required extra steps to create an accurate, georeferenced point cloud. We needed several trips to the site to establish control and targets and scan the bridge arch. Those additional set ups would add time we didn’t have, and range limitations of the scanning technology wouldn’t allow us to shoot across the river.”

Based on the exercise, Foth determined they needed a single instrument that could integrate total station measurements with high-speed, georeferenced 3D scans. After evaluating available technology, they acquired a Trimble SX10 scanning total station. The SX10 combines surveying, imaging and scanning (up to 26,600 points per second with a range of 600m).

“The SX10 was not only going to give us the range we would need, it would allow us to eliminate the need for additional field work because it automatically integrates the project coordinate system and georeferences the points,” says Ben Sullivan, a lead geospatial specialist at Foth. “That would save us significant time both in the field and in processing the data in the office. It also offers a true survey workflow that our crews are already familiar with so we’d need little training – and it would provide a safer working environment.”

Banking on the technology

Based on the CRFCS project schedule, Foth began with Site One, a 1.4-mile-long rugged stretch of river bank about 6 miles north of downtown, where an operational railroad bridge crosses the river and several abandoned, 100-year-old bridge piers dot the water. Foth’s specific assignment was to capture railroad track locations and define the shape and size of the existing and inactive bridge piers to within 0.05 of a foot. That data would be used in a hydraulic model to understand the effects of water flow with those objects in the river.

In the fall of 2017 a two-person crew launched their motorboat carrying a Trimble R10 GNSS rover, the SX10, a Trimble Yuma 2 tablet computer and a Trimble TSC3 data controller. Their first task was to find safe, suitable spots on land to set project control, locations that would also provide good vantage points for scanning. Criss-crossing the 500-ft-wide river, they navigated to nine locations to establish control using the R10.

With survey control completed, they switched to the SX10 and followed the same navigational path. Using the same control points, the team set up the robotic scanner and captured the scene, from the railroad bridge piers all the way downstream to the old pier remnants, collecting features along the river and across the river at distances of nearly 1,000 ft. To gain even more detail and perspective on the scanning objects, the crew performed resections between the control points to acquire several separate 3D measurements of the same feature, enabling them to better show the varied bridge piers’ characteristics. In total, it took them two days to scan the tracks and 12 piers, collect 11.8 million 3D points and take color photos with the built-in camera for coloring the point cloud.

“We were asked to define the size of the piers on one plane,” says Sullivan. “But the old piers had vertical size discrepancies between four and six feet and varied shapes. Accurately measuring those variations with a total station would have been difficult. Scanning was able to quickly and accurately capture the unique detail of each pier and in color, allowing the engineers to better evaluate and incorporate those structures into their model for a more precise analysis.”

Back in the office, they used the drag-and-drop nature of Trimble Business Center (TBC) software to take the data from the TSC3 and Yuma and seamlessly integrate the GNSS and scanning data into one project. In a few hours, they created and delivered a 3D point cloud for the engineering design firm to immediately integrate into its own hydraulic analysis software.

“Originally, the data deliverable request was for a standard CSV text file,” says Budde. “But we knew a point cloud would give them more data depth and much better feature definition for further analysis. After they realized how beneficial the point cloud information would be, they asked for the 3D data going forward, starting with Site Two.”

Racing the Clock

Foth returned to the Cedar River in Spring 2018 to tackle Site Two. This particular 1-mile stretch of riverbank will have a permanent flood wall so Foth needed to collect a topographic survey and as-built data of ground-level and submerged infrastructure and three bridges to inform the design of the wall construction.

Collecting that information, however, required the city to close an upstream dam for the first time in its history. That opened a flood of unknowns, and led the city to err on the side of caution and only allowing the dam to be closed for four hours.

Because of the uncertainties, Foth dispatched two, two-person crews who set ten control points with the R10 and started scanning with the SX10 while the water level was lowering. To accommodate the notably high-accuracy vertical precision required – one specific bridge had a 1.5-in clearance tolerance to the bottom of the bridge – the crews also used a Trimble DiNi digital level at each control point to ensure height elevations would be within the project’s 0.01 of a foot specification.

In parallel with the control crew, the scanning crew used project control to capture 3D points of the infrastructure. Using four separate locations on the east side of the river, they scanned the entire length of the west bank AOI, collecting features up to 800 ft away including bridges and bridge arches, existing flood structures, utilities and submerged features like underground pipe networks as the riverbank became more exposed. They also performed resections between control points to gain better perspective on specific objects such as the bridge structures. For the 12th Avenue Bridge, the structure with the tight 1.5-in clearance design, the Foth team captured additional scans of its underside to capture a 3D point every 0.125 to 0.25 inches. By the time the dam was reopened, the crews had finished the job and collected 27.6 million points.

“Without the speed and scanning range of the SX10, we would not have been able to do this job,” says Ben Sullivan, a lead geospatial specialist with Foth. “Conventional survey technology would’ve required up to three times more man-hours and we would’ve collected only about 20 percent of the data detail we captured with the scanner.” Had Foth used a dedicated scanner, they would have incurred significant additional time setting targets intervisible from the multiple setups.

Similar to Site One, they used the TBC software to integrate the diverse data streams from the R10, DiNi level and the SX10 to process and validate the optical, leveling and GNSS data into one georeferenced project. In processing the point cloud and using the TBC plane-definition and cross-section tools, the Foth team discovered an unexpected data discrepancy for  the 12th Avenue Bridge. Historical data indicated that the lowest arch points were the lowest clearance point. The 3D scanning data clearly showed a support pipe under the bridge deck that was lower than the bridge arches, revealing the pipe as the true low point for vertical clearance.

“The wall structure would be constructed using pre-cast concrete panels fabricated off-site,” says Budde. “If they’d designed and built the panels on the original assumption, they would’ve had a costly mistake.”

Instead, the engineering design firm has a precise 3D point cloud to integrate into its own modeling software, providing the information they need to create the most effective permanent flood wall for that part of the west bank.

With many more phases of the CRFCS to complete, it’s a good bet that Foth will face its share of complexity in the future. Based on the results of its SX10’s scanning debut, they are confident they have the right tools to succeed.

Mary Jo Wagner is a Freelance Writer, Editor, and Media Consultant based in Vancouver, BC. She can be reached at mj_wagner@shaw.ca.