Help


Abreviations

BC: British Columbia

BCSIMS: British Columbia Smart Infrastructure Monitoring System

GSC: Geological Survey of Canada

kSI: Katayama’s Spectral Intensity

IA Stations: Internet Accelograph stations

ITS: Intelegent Transportation System

MOTI: BC Ministry of Transportation and Infrastructure

PGA: Peak ground acceleration

PGV: Peak ground velocity

PGD: Peak ground displacement

PGC: Pacific Geoscience Centre

SHM: Structural Health Monitoring

SGMN: Strong Ground Motion Network

STA/LTA: Short Time Average to Long Time Average

UBC: University of British Columbia

USA: United States of America

USGS: United States Geological Survey

VIF: Virtual Input File

INTRODUCTION

Managed by the MOTI, the BCSIMS is a province-wide seismic-focused monitoring network, which encompasses a comprehensive SHM system and a SGMN. The BCSIMS is an ITS system for smart remote monitoring of structures and ground motions that is used to help ensure highway structures safety and for post-seismic and other emergency responses.

  1. Provide a real-time seismic structural response system to enable rapid deployment and prioritized inspections of MOTI’s structures
  2. Develop and implement a SHM program to address the need for safe and cost-effective operation of structures in BC

The SHM system is a comprehesive set of tools that aims to provide quantitative and reliable data on the real conditions of structures, observe its evolution and detect the appearance of degradations; therefore, it is possible to obtain a real-time picture of the structures’ state and evolition. As part of the SHM system, seismic instruments have been installed on selected infrastructure (e.g., bridges on lifeline routes) in BC to continuously monitor the seismic vibrations on these structures. In adition to the seismic vibrations, variation of temperature, wind, pore pressure, strain has also been measured on structures. The objective of the SHM system is to provide the MOTI (also the public) with the timely information about the performance of the instrumented structures immediately following a significant event such as a strong earthquake or a significant wind. Such information is then used by the MOTI to help assess the safety of these structures and to support cost-effective inspection, maintenance and renewal programs. The SHM system currently includes 14 bridges and 1 tunnel in BC. The number of structures with SHM installed in BC is expected to increase in the upcoming years.

The SGMN system consists of a network of earthquake instruments installed close to earthquake epicenter in urban centers to measure and report on earthquakes with amplitudes capable of causing damage. The SGMN in BC has been maintained by the GSC and the MOTI. The SGMN is a network of approximately 160 acceleration sensors installed across BC. These sensors are permanently installed in locations such as public schools, government offices, fire halls, BC ambulance stations, bridges, etc. They monitor and record seismic activities in real-time at and around their installed location. Once trigged by an earthquake, these sensors will immediately report the level of shaking to the BCSIMS data center. A shake-map and an earthquake report will automatically be generated following the earthquake. The shake-map is an interactive color-coded map generated using the recorded ground motion data from the acceleration sensors that shows the estimated level of shaking across the earthquake affected region. The earthquake report includes additional information about the earthquake such as the epicentre, magnitude, estimated level of shaking at the location of bridges, schools and tunnels, etc.

The BCSIMS incorporates the SGMN and SHM system into an automated and user-interactive online platform (http://www.bcsims.ca/ ). An earthquake notification system and web pages have been developed for the BCSIMS to provide the MOTI and the public with detailed information about the seismic activities in and around BC.

STRONG GROUND MOTION NETWORK

Earthquake Monitoring

The BCSIMS network maintains a continuous online link to the USGS to obtain the current earthquake activities in Canada as well as in the USA states of Washington, Oregon and Alaska. These earthquake activities are immediately posted on the BCSIMS webpages (http://www.bcsims.ca/) with the following meta data:

  • Earthquake magnitude,
  • Epicentral location
  • Hypocentre depth
  • Date & time of the earthquake
  • Region

An automated service sends out immediate earthquake notifications by e-mail to registered users of the BCSIMS network.It contains the meta data of the earthquake as well as any earthquake report generated by the BCSIMS. The registration for the earthquake notification service is open to public from the BCSIMS webpage.

Internet accelerometers (IA) network

The SGMN consists of approximately 160 IA Stations which are deployed across the province of BC (Rosenberger et al, 2004). The IA Stations are designed to trigger and record earthquake ground motions in real-time and send the recorded earthquake data along with calculated seismic parameters over the Internet to a central data center. The communication between each IA Sensors and the data center is done via four relay servers, which are located at UBC in Vancouver; PGC in Sidney on Vancouver Island; Kamloops, BC; and in Ottawa, Ontario. This spread of geological locations of the servers increases the robustness of the entire SGMN.

Each IA Station has a digital recorder built internally that allows at least three days of raw data to be stored locally in a ring buffer of the sensor, and it also allows the calculation of the following earthquake parameters by the sensor:

  • Peak ground acceleration (PGA)
  • Peak ground velocity (PGV)
  • Peak ground displacement (PGD)
  • Katayama’s Spectral Intensity (kSI) (Katayama et al. 1998)
  • STA/LTA ratio (Short Time Average to Long Time Average)
  • Spectral acceleration at least at two periods such as 0.2 seconds and 1.0 second

A smart algorithm embedded in the IA Station triggers the sensor to earthquake motions and stores the event data. This algorithm is based on the combination of the threshold values for these earthquake parameters. When triggered, the sensor immediately logs the earthquake record as an event file on the sensor memory, and it then sends out the first message to the BCSIMS data center. Upon de-trigger, a second message is sent out to the data center that contains the calculated seismic parameters.

Shake-maps

A shake-map is a color-coded map that shows the estimated distribution of the strong ground motion parameters (e.g., PGA or earthquake intensity) over an earthquake affected region. They provide vital information about the potential for damage from the earthquake. These shake-maps, therefore, could be used by federal, provincial, and local organizations, both public and private, for post-earthquake response and recovery as well as for preparedness exercises and disaster planning. This enables emergency responders, emergency coordinators, and inspection and maintenance personnel to quickly assess the shaking levels across the affected areas and at the location of critical infrastructure. It also allows these agencies to prioritize and maximize the effective use of their resources after an earthquake.

The calculation of the shake-map includes the Ground Motion Prediction equations developed by Boore and Atkinson 2008 and Atkinson and Boore 2011. A shake-map will be generated automatically by the BCSIMS following moderate or large earthquakes if the following criterions are met:

  • The epicenter of the earthquake is less than 200 km from the nearest IA station
  • The magnitude of the earthquake is bigger than 3
  • At least one IA station is triggered

Shake-maps are posted on the BCSIMS web pages, and they can be viewed with different superimposed layers showing the locations of infrastructures such as bridges, buildings, tunnels or schools. The shake-maps are generated and posted on the BCSIMS web pages approximately 30 minutes on average following an earthquake.

Earthquake Report

Upon the generation of a shake-map, an earthquake report is automatically generated and e-mailed to a predefined subscriber list such as bridge inspection engineers. This report is also publicly available via the BCSIMS web pages. The earthquake report includes the following:

  • Snapshot of the shake-map
  • The meta data of the earthquake (e.g., location, magnitude, depth, etc.).
  • The population count of the MOTI structures such as bridges, tunnels, etc. around the epicentre
  • List of MOTI structures with estimated level of earthquake shaking
  • List of the IA Stations that are triggered by that earthquake and their peak responses.

The data recorded by each triggered IA Station is automatically analyzed at the data center, and the results are also included in this report. The analysis results include recorded acceleration, calculated velocity and displacement time histories as well as the acceleration response spectrum and the smoothed Fourier Amplitude Spectrum.

STRUCTURAL HEALTH MONITORING NETWORK

The practical significance of the seismic-focused SHM cannot be overstated; for example, after significant large earthquakes, it is common to see damaged infrastructure. The damaged structures typically require detailed and time-consuming inspections to decide whether the damage is structural or not, and if the damage is of a structural nature, then decide if it is safe to re-occupy and use them. This process can result in heavy financial losses for the owners of such structures. The SHM systems in the BCSIMS network provides the means to help make such decisions faster and with more confidence.

In parallel to recent developments in sensor and recording technologies, 14 bridges and 1 tunnel as seen in Table-1 have been installed with SHM systems. These structures have been instrumented with earthquake sensors that continuously measure the level of the vibration on these structures during an earthquake. These earthquake sensors can also measure day to day ambient vibrations which can then be used to monitor for other changes in the behaviour that could indicate structural damage. The structures have also been instrumented with other types of sensors to monitor other movements and environmental conditions. The type and number of sensors are different at each structure. These SHM sensors are remotely accessible via the Internet, and they continuously stream real-time structural vibration data to the remote data center for analysis.

Table-1: List of structures with SHM systems installed and currently being monitored in real-time in the BCSIMS

No Structure name Year Installed Type of sensor Number of channels
1 Pitt River Bridge (PRB) 2009 A W 46
2 Kensington Avenue Underpass (KAU) 2013 A T H 30
3 Gaglardi Way Underpass (GWU) 2013 A T H 22
4 176th Underpass (176U) 2013 A T H 26
5 Fraser Heights - Wetlands (FHB) 2013 A T H 20
6 Queensborough Bridge (QB) 1996 A P 12
7 George Massey Tunnel (GMT) 1996 A P 11
8 Ironworkers Memorial Second Narrows Crossing (IMSNC) 2011 A S W T 121
9 Port Mann Bridge (PMB) 2013 A W D T H P 336
10 William R. Bennet Bridge 2008 A 12
11 French Creek Bridge 1997 A 12
12 Portage Creek Bridge 1983 A D 41
13 8264 BNSP Bridge 2014 A D T H 36
14 8270 BNSF Viaduct East Mill Access 2014 A D T H 84
15 8313 Hwy-17 Deltaport 2014 A D T H 36

A: Acceleration sensor, W: Wind sensor, P: Pore pressure, H: Humudity, D: Dipsplacement sensor, T: Temperature sensor, S: Strain Guage

The collected data from each structure is archived and analysed in real-time in the data center, and the data analysis involves the following:

  • Data archiving using VIF format
  • Statistical variations of structural vibrations
    • Maximum
    • Minimum
    • Mean
    • Standard deviation
    • Skewness
    • Kurtosis
  • Dynamic characteristics of the structure
    • Modal frequencies
    • Modal Damping ratios
    • Mode shapes
  • Drift calculation

These analysis results are done automatically in real-time with no human interactions. They are then used to detect damage on structures, to make fast decisions on the safety level of the structure, and the actions that need to be taken following a significant earthquake.

Data archiving

Unique data archiving standards and protocols have been developed for the BCSIMS. The advantage of this approach is that it enables consistency and platform-neutrality across all data acquisition systems and hardware platforms, thereby simplifying the downstream processing.

The collected data from each structure is streamed via the Internet to a data center where the data is stored in 5-minute lengths of VIF files. The VIF file is a binary file that contains raw data from all the SHM sensors. The length of the VIF file is user adjustable, and it is set to a 5-minute length by default because this length enables the data analysis server to keep up with the post processing of the data from all the structures simultaneously. As soon as a new VIF file becomes available in the data center, it is further compressed so as to minimize the disk space.

Drift analysis

Structural damage at bridge piers is often directly related to displacement demands. It is usually controlled in many bridge codes by imposing displacement (or drift) limits on these structural members. The drift at a bridge pier is defined as the relative displacement of the pier top with respect to its base. To calculate the drift in BCSIMS, the displacements at these two locations are estimated from recorded acceleration data at the top and the base of the pier. This estimation of displacements from acceleration data is done by double integration of the recorded accelerations after appropriate filtering. The peak drift values during an event are stored in the database. Any drift value exceeding a predefined threshold value may be indicative of possible damage in the structure, and if this is the case, an e-mail notification is sent out to a predefined list of users.

References

Atkinson, G.M., and Boore, D.M. 2011. Modifications to existing ground-motion prediction equations in light of new data. Bulletin of Seismological Society of America, 101(3): 1121–1135.

Boore, D.M., and Atkinson, G.M. 2008. Ground-motion prediction equations for the average horizontal component of PGA, PGV, and 5%-Damped PSA at spectral periods between 0.01 s and 10.0 s. Earthquake Spectra, 24(1): 99–138

Katayama, T., Sato, N., and Saito, K. 1998. SI-Sensor for the identification of destructive ground motion. In Proc. Ninth World Conference of Earthquake Engineering, Tokyo-Kyoto, Vol. VII, pp. 667–672.

Kaya, Y., and Safak, E. 2014. Real-time analysis and interpretation of continuous data from structural health monitoring (SHM) systems. Bulletin of Earthquake Engineering, 13(3): 917–934. doi:10.1007/s10518-014-9642-9.

Kaya Y, Ventura C, Huffman S, Turek M (2017) British Columbia smart infrastructure monitoring system. Canadian Journal of Civil Engineering 44(8):579–588. https://doi.org/10.1139/cjce-2016-0577

Rosenberger, A., Beverley, K., and Rogers, G. 2004. The new strong motion seismic network in southern British Columbia, 13th World Conference on Earthquake Engineering Vancouver, B.C., Canada. Paper No. 3373

USGS, United States Geological Survey, Available from http://earthquake.usgs.gov