1. Introduction
Continuous and systematic monitoring of the displacements of infrastructures plays an important role in its maintenance and operation, especially for detecting vulnerable conditions to provide adequate remedial measures at early stages (Del Soldato et al., 2016). Bridges are one such type among various infrastructures, which undergoes both bridge settlement and structural deflection. Bridge settlement is caused when the structure increases the effective stress on the soil that leads to soil consolidation. Eventually, the bridge will vertically be displaced from its originally established ground level and the excessive differential settlement of bridge foundations will cause undesirable damagesto the bridges(Parks et al., 2018).At the same time, the dynamic responses of the bridge structure over a period of time, such as vertical deflection and horizontal displacements need to bemonitored to assess the structuralstability ofthe bridges.In-contactsensors like accelerometers, Global Navigation Satellite System (GNSS) have demonstrated as the reliable inspection technique forstructural behaviors, and in-situ techniques including leveling and Ground Based – Synthetic Aperture Radar Interferometry InSAR (GB-InSAR) showed the capability for measuring the deformation of bridges (Yi et al., 2013). However, these methods are often expensive and time-consuming due to inadequate manpower for a large number of bridges.
InSARisidentified as a powerful technique to detect and monitoring displacements of infrastructure and slow deformation overlarge areas(Ferretti et al., 2001). Time-seriesInSAR techniques, for instance, Persistent Scatterer Interferometry (PSI) (Ferretti et al., 2001), Stanford Method for Persistent Scatterers (StaMPS) (Hooper et al., 2004) andSmallBaselinesSubset(SBAS) (Berardino et al., 2002) have proved its capability to measure the displacements at sub-centimeter level accuracy.
In this study, a StaMPS time-series PS-InSAR technique is applied to assess the structural behavior of the case study bridges, Kimdaejung bridge, and Deokyang bridge located in the Jeollanam-do, Korea. To achieve this, we acquired Cosmo-SkyMed X-band SAR data over the study area during the period 2013 to 2019. Besides, we aim to address the geolocation error that occurred along the bridge in the postprocessing stage.
2. Study Area
The Jeollanam-do province is located in the southwestern of the Korean peninsula, in which two bridges were selected for the monitoring of long-term bridge displacements using time-series SAR interferometry namely, Kimdaejung bridge and Deokyang bridge (Fig. 1).The Kimdaejung bridge is an extradosed bridge which is a type of cable-stayed bridge having a total length and width of 925 m and 24 m, respectively. (Jung et al., 2019)reported the first long-termdeflection signal in the Kimdaejung bridge from 2013 to 2017 using the time-seriesInSARtechnique.The Deokyang bridge is a prestressed concrete box bridge having a length and width of 450 m and 35.8 m, respectively. The schematic representation of the case study bridges is illustrated in Fig. 2.
Fig. 1. Cosmo-SkyMed SAR data used in this study (a) Mokpo (b) Yeosu.
Fig. 2. Schematic representation of the case study bridges (a) Kimdaejung Bridge (b) Doekyang Bridge.
We acquired X-band Cosmo-SkyMed (CSK) SAR data over Kimdaejung bridge (Mokpo) and Deokyang bridge (Yeosu)from2013 to 2019.CSK SARdata have the advantage of high spatial resolution with a pixel spacing of 1 m x 2.2 m forrange and azimuth direction, respectively.TheCSK sensorspecification isillustrated in Table 1.
Table 1. Description of Cosmo-SkyMed SAR data used in this study
3. Methods
1) Time-series InSAR workflow
StaMPS PS-InSAR technique was applied in this study to monitor the displacement of the selected bridges (Hooper et al., 2004). StaMPS - PSI relies on the man-made orstable natural targets called Persistent Scatterers (PS) which show a reliable and dominant scatteringcharacteristic andcoherence remainsunchanged over the period (Ferretti et al., 2001; Hooper et al., 2004). Fig. 3 illustrates the workflow that applied in this study.
Fig. 3. StaMPS PSI workflow applied in the study and post-processing methodology for geolocation error.
The workflow begins with the Coregistration of SLC images with reference to the single master scene selected based on the minimum geometric and temporal decorrelation. Fig. 4 shows the plot between perpendicular baselines versustemporal baselines with respect to themasterscenes 2015/11/21 and 2016/05/27 for Kimdaejung and Deokyang bridge, respectively. Following that, we implemented InSAR Scientific ComputingEnvironment(ISCE) package for generating the stack of interferograms (Agram et al., 2013). The interferogramswere generated asfullresolution products (i.e., 1 × 1 look), therefore there is no smoothing filter applied. In order to remove the contribution of the topographic phase from the interferometric phase, we used ShuttleRadarTopographicMission (SRTM) DEM data having 30 m spatial resolution (Farr et al., 2007).
Fig. 4. Perpendicular baseline vs Temporal baseline plot (a) Mokpo (Kimdaejung bridge) (b) Yeosu (Deokyang bridge).
StaMPS PS-InSAR algorithm was applied to the differential interferograms stack for estimating the time-series displacements (Hooper, 2005). The initial persistent scatterers were selected based on the Amplitude Dispersion Index (ADI) < 0.4. Following that, the PS pixels were furtherrefined based on spatial coherence > 0.3. Then, the 3D unwrapping method is carried out in space and time domain (Chen andZebker, 2002; Hooper and Zebker, 2007) which comprise of phase due to displacement, topographic heightresiduals, and atmospheric noise.
2) Geolocation correction
As mentioned earlier, we have used SRTM DEM in this study that has the topographic information (i.e., elevation) only at the time of acquisition.Therefore, the terrain information of the infrastructures constructed after the acquisition of SRTM DEM will not be available. As a result, the incorrect height information willlead to the geolocation offset ofPSpixels, especially for the vertical structures. By estimating the accurate residual topographic height, we can approximately correct the geolocation error of the PS pixels (Jung et al., 2019).The phase due to the topographic component was estimated using equation (1). The PS pixels of the Kimdaejung bridge affected by inaccurate topographic height is illustrated in Fig. 5. Fig. 5(a) shows the zeroelevation information of the Kimdaejung bridge and Fig. 5(b) shows the geolocation of the PS pixels with corresponding residual topographic height (DEM error).As discussed in (Jung et al., 2019), we corrected the geolocation offset of the PS pixels along the bridge as shown in Fig. 5(c).
Fig. 5. (a) SRTM height used in the StaMPS PS-InSAR processing (b) Spatially correlated DEM error (c) Geolocation corrected PS pixels.
\(\Delta \emptyset_{\text {topo }}=\frac{4 \pi}{\lambda} \frac{B^{\perp}}{R \operatorname{Sin} \theta} \Delta H\) (1)
where Δøtopo is topographic phase component in the interferometric phase that estimated by perpendicular baseline B⊥, wavelength λ, and residual topographic height ΔH.
Then, the residual topographic height estimated along line-of-sight direction was derived into offset distance in the horizontal plane (2) and vertical plane (3) using incidence angle θ, and heading angle α. The geolocation corrected PS pixels is illustrated in Fig. 5(c).
Δx = ΔH.cotθ. cos α (2)
Δy = ΔH.cotθ. sin φ (3)
4. Results and Discussions
The long-term time-series displacements of the two bridges processed during 2013 to 2019 are presented in this section.
1) Kimdaejung Bridge
In this study, more than 3900 PS pixels were identified on the Kimdaejung bridge.Themean velocity and time-series displacements were estimated for each PS pixels. Fig. 6 shows the mean subsidence velocity map of Kimdaejung bridge from 2013 to 2019 and the bridge displacement rate varies from -9 mm/year to 4 mm/year. Jung et al. (2019) estimated the long-term deflection of Kimdaejung bridge during 2013 – 2017 and we further continue to monitorthe deflection ofthe bridge as a continuous assessment to preventstructural failure.
Fig. 6. The linear velocity of PS pixels in the LOS direction (a) with geolocation error (b) after geolocation corrected.
The significant downward deflection is determined at the mid-span of the Kimdaejung bridge, whereas the considerable upward deflection is recognized at the towers (Fig. 6). In addition, the geolocation of PS pixelsis now reconstructed at the appropriate location. The time-series displacement at the mid-span of the Kimdaejung bridge is shown in Fig. 7, shows the range between 25mmto -35mmfor 7 years.The spatial distribution of displacementrate along the longitudinal axis bridge is presented in Fig. 8.Besidesthe mid-span and tower, the Kimdaejung bridge presents a near-zero displacement rate.
Fig. 7. Time-series line-of-sight displacement at the mid-span of the Kimdaejung bridge.
Fig. 8. Spatial distribution of mean velocity rate of PS pixels along the Kimdaejung bridge.
2) Deokyang Bridge
The mean velocity and time-series displacement of Deokyang bridge is presented in Fig. 9 between 2013 and 2019. Forthe first time, we report the displacement of the Deokyang bridge. Unlike Kimdaejung bridge, the Deokyang bridge does not have any towers, therefore we observed constant LOS displacement rate in the center span of the bridge. The geolocation correction of the PS pixels is well agreed with the optical image of Deokyang bridge (Fig. 9(b)). The bridge displacement rate at Deokyang bridge ranges between the -2 mm/year and 2 mm/year.
Fig. 9. Mean velocity of PS pixels over Deokyang bridge (a) with geolocation error, (b) geolocation corrected (c) timeseries displacement at mid-span of the bridge.
5. Conclusions
In this study, we monitored the displacement of the selected bridges in the South Korea 1) Kimdaejung bridge and 2) Deokyang bridge using StaMPS timeseries PS-InSAR analysis. The mean displacement rate of Kimdaejung bridge and Deokyang bridges are -8.8 mm/year and -1.34 mm/year, respectively. This study corroboratesthe utilization ofspace-borne InSAR technique for reliable and continuous monitoring of bridge displacements.Thisresultantspatial and temporal bridge displacements will provide crucial information for preventing the structural failure of the bridges.
Acknowledgements
This work was supported by MSIT (Ministry of Science and ICT), Korea under the ITRC (Information Technology Research Center) support program (IITP2020-2018-0-01424)supervised by the IITP(Institute of Information&CommunicationsTechnologyPromotion).
참고문헌
- Agram, P. S., R. Jolivet, B. Riel, Y. N. Lin, M. Simons, E. Hetland, M. P. Doin, and C. Lasserre. 2013. New Radar Interferometric Time Series Analysis Toolbox Released, Eos, Transactions American Geophysical Union, 94(7): 69-70.
- Berardino, P., G. Fornaro, R. Lanari, and E. Sansosti. 2002. A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms, IEEE Transactions on Geoscience and Remote Sensing, 40(11): 2375-2383. https://doi.org/10.1109/TGRS.2002.803792
- Chen, C. W. and H. A. Zebker, 2002. Phase unwrapping for large SAR interferograms: Statistical segmentation and generalized network models, IEEE Transactions on Geoscience and Remote Sensing, 40(8): 1709-1719. https://doi.org/10.1109/TGRS.2002.802453
- Del Soldato, M., R. Tomás, J. Pont Castillo, G. Herrera García, J. C. García Lopez-Davalillos, and O. Mora, 2016. A multi-sensor approach for monitoring a road bridge in the Valencia harbor (SE Spain) by SAR Interferometry (InSAR), Rendiconti Online Societa Geologica Italiana, 41: 235-238 https://doi.org/10.3301/ROL.2016.137
- Farr, T. G., P. A. Rosen, E. Caro, R. Crippen, R. Duren, S. Hensley, M. Kobrick, M. Paller, E. Rodriguez, L. Roth, D. Seal, S. Shaffer, J. Shimada, J. Umland, M. Werner, M. Oskin, D. Burbank, and D. Alsdorf, 2007. The Shuttle Radar Topography Mission, Reviews of Geophysics, 45(2).
- Ferretti, A., C. Prati, and F. Rocca, 2001. Permanent scatterers in SAR interferometry, IEEE Transactions on Geoscience and Remote Sensing, 39(1): 8-20. https://doi.org/10.1109/36.898661
- Hooper, A., H. Zebker, P. Segall, and B. Kampes, 2004. A new method for measuring deformation on volcanoes and other natural terrains using InSAR persistent scatterers, Geophysical Research Letters, 31(23).
- Hooper, A. J., 2005. Persistent scatter radar interferometry for crustal deformation studies and modeling of volcanic deformation, Journal of Geophysical Research, 112(B07407): 1-21.
- Hooper, A. and H. A. Zebker, 2007. Phase unwrapping in three dimensions with application to InSAR time series, JOSA A, 24(9): 2737-2747. https://doi.org/10.1364/JOSAA.24.002737
- Jung, J., D.-J. Kim, S. K. Palanisamy Vadivel, and S. Yun, 2019. Long-Term Deflection Monitoring for Bridges Using X and C-Band Time-Series SAR Interferometry, Remote Sensing, 11(11): 1258. https://doi.org/10.3390/rs11111258
- Parks, S., M. Yang, S. Gajan, and Q. Pei, 2018. Strengthbased differential tolerable settlement limits of bridges, Advances in Structural Engineering, 21(1): 46-58. https://doi.org/10.1177/1369433217706779
- Yi, T.-H., H.-N. Li, and M. Gu, 2013. Experimental assessment of high-rate GPS receivers for deformation monitoring of bridge, Measurement, 46(1): 420-432. https://doi.org/10.1016/j.measurement.2012.07.018