Using GPR to estimate wall thickness of bridge structures
close X
Nav Menu

Using GPR to estimate wall thickness of bridge structures


he Saratoga Creek Bridge along State Route 9 (Figure 1) was constructed in 1902 and enables the safe and stable connectivity between the City of Saratoga the community of Felton in California. The bridge has a two-span, earth-filled, concrete arch, with rubble masonry spandrel1 walls. Previous surveys found no evidence of reinforcing steel bars at the bridge abutments2 or at the pier3. These structural deficiencies, coupled with mortar joint deterioration, makes the bridge susceptible to damage during a seismic event, particularly considering its proximity to the San Andreas fault system, located approximately half a mile away.

Figure 1
State Route 9 Saratoga Creek Bridge Project Location.

Due to these issues, the bridge needs to be replaced or extensively renovated. The chosen way forward is to have a “hybrid” bridge design, where a new steel girder bridge will be built within the body of the existing bridge with the existing masonry walls and stone arches serving as a façade concealing the new support columns. The design retains the look and feel of the existing stone bridge (which has strong local support) while also ensuring the structure will survive a seismic event and is the fastest method to complete construction.

To assist in the hybrid bridge design, the Geophysics and Geology Branch (GGB) of Caltrans (California Department of Transportation) was tasked to identify the construction details for the concrete arch and rubble masonry spandrel walls of the bridge structure in January 2020. The bridge structure had been successfully surveyed with a GPR in 2010 so GPR was chosen for this non-destructive evaluation project.

The goals were to find the thickness of the concrete arch at the crown and base, the thickness of masonry rubble spandrel wall and the depth of the concrete pier.

Survey & Results

For the wall survey, unidirectional profiles, starting from the top, were acquired vertically with pulseEKKO® PRO 500 MHz center-frequency transducers, mounted on a custom frame with access from the bridge deck (Figure 2, left). To access the more difficult to reach parts of the arches and walls, a snooper truck was used (Figure 2, right). Multiple lines were also collected on the base and crown of the concrete arch of the bridge.

Figure 2
GPR line data collection using a pulseEKKO® PRO 500 on the bridge abutment (left) and Conquest® 100 on the concrete arch at Pier 2 (right).

Figure 3 shows all the GPR lines collected along the non-reinforced concrete arches of the bridge, the center column, and the spandrel walls of Pier 2 and Abutment 3.

As GPR work was limited to two days for this survey, all data were collected and saved in different projects on the GPR system and processed in EKKO_ProjectTM GPR software to obtain estimated thickness measurements. To get an accurate GPR wave velocity estimate, a core hole was drilled at the base of the concrete arch at Abutment 3 (Figure 3, right) to measure the bridge wall thickness. The GPR wave velocity was calculated by correlating the wall thickness from the core to the time of the GPR reflection from the back of the wall. This velocity was assumed to be representative of the rest of the concrete wall.

Figure 3
GPR line data collected on the Saratoga Bridge at labeled locations with estimated wall thicknesses. Solid lines represent data collected on the north wall and dotted line is data collected on the south wall. Thickness measurements down Abutment 3 indicated the wall thickness increased towards the base.

Interpreting and finding the wall thickness from the GPR data was simple, given that a good dielectric contrast existed between the sandstone block wall and the packed earth between the bridge walls. The GPR cross-sections show the transition from uniform to scattered radar reflections that was interpreted as the transition from the wall to rubble fill (Figure 4).

Apart from successfully measuring the wall thickness of the concrete arch at the base and crown from the GPR data, they were also able to confirm the existence of batter (an inclined slope) on the wall at Abutment 3, as the estimated wall thickness appeared to increase towards the base of the wall (Figure 4, right). The results at Pier 2 provided more insights, with the GPR data indicating a relatively constant thickness at the center column and spandrel wall (Figure 4, left), implying the absence of wall batter. It also highlighted the significantly thinner spandrel walls when compared to Abutment 3 (Figure 3, left).

Figure 4
GPR cross-sectional data showing the approximate mean column thickness (4.12 ft) observed on the North Wall Center Column above Pier 2 (Left) and varying wall thickness at Abutment 3 confirming the existing of batter (Right).


GGB received positive feedback from the bridge design team that the GPR interpretations provided valuable details to draft their final design for preserving the existing bridge. The hybrid “bridge within a bridge” design was accepted, and construction is planned to start in September 2022.

Data courtesy of Bill Owen from Caltrans.


1 A spandrel is the triangular space formed by an arch, the top or deck of the bridge, and any vertical structural elements.
2 A bridge abutment is the part of the bridge foundation that rests on the ground at either end of the bridge.
3 A pier is the main support column for the span of the bridge deck that crosses between abutments.

Contact Us