Conference Agenda

Natural Resources Canada (and its predecessors) has been responsible for six generations of seismic hazard maps for Canada, starting in 1953. These have formed the basis for the seismic design provisions of the National Building Code of Canada, and thus contribute to the seismic resistance of Canada’s building stock. This paper indicates how the hazard estimates have changed since 1953. A simple model for possible change suggests that some countries start high and in successive hazard estimates approach the target (“true”) value from above, but most countries start low and approach the value from below. In either case the hazard estimate seldom moves monotonically towards the target level, but oscillates above and below the trend. This paper gives a 65-year history of 5% damped spectral acceleration at 0.2 s estimate for the 2%/50yr probability level for Montreal, Vancouver and Victoria. Site Class C was used for the comparison. The Sa(0.2) parameter was not available until 2005 and neither were 2%/50yr probability estimates. Therefore, the earlier short-period hazard estimates were adjusted to be equivalent to Sa(0.2) at the 2%/50yr probability level by using ratios of results from the 4th Generation model. Hazard estimated for Montreal and Vancouver have trended up reasonably steadily at about 0.6% per year while for Victoria the increase has been faster and chiefly occurred in 1985 and 2020. While changes in the seismic hazard estimate reflect evolution of the seismic hazard models, they are only one component controlling changes in the base shear. Also, introduction of other code restrictions, application of better analysis methods, and improvements in associated standards mean that today’s buildings are more earthquake resistant than those built 65 years ago.

The 2015 National Building Code of Canada (NBCC) uses hazard results calculated using the GSC’s 5th Generation hazard model, but uses just a single value, the mean. However, the mean hazard value has uncertainty associated with it, and the amount of uncertainty has implications for the reliability of engineering designs. Uncertainty arises from aleatory and epistemic uncertainty in the model. The uncertainty spread is illustrated through the use of percentiles (5, 16, 50 (median), 84, 95, etc; also called fractiles) of the distribution. The mean hazard values, as used in NBCC2015, correspond to the 60-85th percentiles of the distribution. Calculated distributions are provided for La Malbaie, Montreal, Toronto, Tofino, Victoria, Vancouver, and Kelowna to compare their spreads. The spreads are large, and for most localities can be approximated to a lognormal distribution. The slope of the cumulative distribution (proportional to standard deviation) can be similar (or different) for all periods at a single site, but differs greatly between sites. It is largest for low-seismicity eastern sites and smallest for sites above the Cascadia subduction zone. The main contributors to the spread is the uncertainty in the Ground Motion Models (for most places in Canada) and the uncertainty in the earthquake rates for low-hazard regions.

Canada's 6th Generation seismic hazard model has been developed to generate seismic design values for the 2020 National Building Code of Canada (NBCC2020). The model retains most of the seismic source model from the 5th Generation, but updates the earthquake sources for the deep inslab earthquakes under the Straits of Georgia and adds the Leech River Valley - Devil’s Mountain faults near Victoria. The rate of Cascadia megathrust earthquakes is also increased to match an improved paleoseismic record. Two major changes in the ground motion model (GMM) are A) adoption of modern Ground Motion Models (GMMs), together with a classical weighted-GMM approach replacing most of the three-branch representative suites used in NBCC2015. and B) direct calculation of hazard on various site classes using representative Vs30 values, rather than provision of hazard values on a reference Class C site and then applying F(T) factors. Computations are now being performed with the OpenQuake engine, which has been validated through the replication of the 5th Generation results. Seismic design values (on various Soil Classes) for PGA, and for Sa(T) with T = 0.2, 0.5, 1.0, 2.0, 5.0 and 10.0 s are proposed for NBCC2020 mean ground shaking at the 2% in 50-year probability level. The paper discusses chiefly the change in Site Class C values relative to 2015 in terms of the changes in the seismic source model and the GMMs, but the changes in hazard at other site classes that arise from application of the direct-calculation approach are also illustrated.

Canada's 5th Generation seismic hazard model forms the basis for the seismic design provisions of the 2015 National Building Code of Canada (NBCC). We deaggregate the seismic hazard results for selected cities to help understand the relative contributions of the earthquake sources in terms of distance and magnitude. Deaggregation for a range of probabilities and spectral accelerations (Sa) from 0.2 to 10.0 seconds is performed to examine in detail the hazard for two of Canada's largest urban centres at highest risk, Vancouver in the west and Montréal in the east. A summary table of deaggregated seismic hazard is provided for other selected Canadian cities, for Sa(0.2), Sa(2.0) and peak ground acceleration (PGA) at a probability of exceedence of 2%/50 years. In most cases, as the probability decreases, the hazard sources closer to the site dominate. Larger, more distant earthquakes contribute more significantly to hazard for longer periods than shorter periods. The deaggregations allow better-informed choices of scenario events and for the selection of representative time histories for engineering design.

The 6th Generation seismic hazard model of Canada is being developed to generate seismic design values for the 2020 National Building Code of Canada (NBCC2020). Ground-motion models (GMMs) from the Next Generation Attenuation (NGA)-West 2 and NGA-East programs are used, and epistemic uncertainty in ground-motion models is captured through the use of a classical weighted logic tree framework. For the first time in Canada, seismic hazard is computed directly on primary (e.g. AE) seismic site classes from their time-averaged shear wave velocities in the upper 30 m of the crust (VS30). This approach simplifies the way end users will determine seismic design values for a given location and site class, while having other technical advantages such as capturing epistemic uncertainty in site amplification models. It will remove the need for separate site amplification look-up tables in the building code, enabling users to simply supply their location and site class to determine seismic design values. In general, the new ground-motion models predict higher hazard in most Canadian localities due to a variable combination of changes in median ground motions, site amplification and aleatory uncertainty.

Strong motion monitoring continues to evolve rapidly in Canada, with many organisations contributing data. This article summarises both the current state of strong motion monitoring across Canada and recent (since 2014) strong motion datasets. As of late 2018, an upgrade of the Canadian National Seismograph Network (CNSN) is nearing completion, resulting in one of the most significant changes in strong motion monitoring in Canada since the first accelerometers were deployed in 1963. As a part of this upgrade, ~100 new strong motion instruments (Nanometric Titans) were deployed (co-located with broadband seismometers) at bedrock sites in the higher seismic hazard regions of Canada. In addition, ~40 stand-alone strong motion instruments will be deployed in the coming year. Currently, Natural Resources Canada (NRCan) continues to operate nearly 100 strong motion Internet Accelerometers (IA’s) across Canada, primarily located on soil sites, and in the urban centres of high seismic risk in southwest British Columbia and southwestern Quebec/eastern Ontario. BC Hydro operates more than 70 strong motion instruments at dam sites and substations across BC. Other strong motion instruments in western Canada are owned by utilities or transportation organisations (BC Ministry of Transportation and Highways has deployed nearly 100 instruments to monitor bridges and critical infrastructure). Ocean Networks Canada now has 5 strong motion instruments and 3 tiltmeters on the seafloor west of Vancouver Island and 15 strong motion instruments onshore Vancouver Island, and UBC and BCSIMS have deployed dozens of instruments, primarily in southwest BC. In eastern Canada, several organisations operate strong motion instruments, including: Hydro-Quebec at 12 dams and substations; Ontario Power Generation and New Brunswick at their nuclear power stations; PWGSC at Parliament Hill; and Gaz Metropolitain at its Montreal LNG plant. Since 2014, more than 139 accelerograms (west only) have been recorded across Canada, mostly in the active tectonic region of Vancouver Island (e.g., recordings of a Mw 4.7 near Victoria). While no large earthquakes were recorded (the strongest ground motions are accelerations of ~ 5% g) these datasets are still useful for comparison with proposed attenuation relations, and for evaluating local earthquake site response. These studies are valuable to engineers evaluating strong ground shaking during future earthquakes.