The consistency of earthquake data and plate tectonic models and other estimates of fault slip, may be tested by estimating rates of motion from the earthquakes using the concept of seismic moment. The contribution of individual events may simply be summed, but a generally better estimate of long-term average slip rate is obtained by integration over magnitudefrequency of occurrence relations. Estimates of fault motion rates associated with earthquakes are possible within about a factor of 2 using this approach if the major sources of uncertainty are given careful consideration; e.g. incompleteness and inaccuracies in the earthquake data; empirical moment-magnitude relations and the effect of their stochasticity; fault 'widths' or depth extents; the recurrence relations; maximum magnitudes and the form of truncation at the maximum magnitude. A formulation for the recommended magnitude density truncation is developed. Application of the latter method to the earthquake data of the offshore transform faults of the Juan de Fuca ridge system, the Queen Charlotte fault zone and the northern Vancouver Island area in each case gives good agreement with rates from plate tectonic models. For the southern San Andreas fault and Gulf of California area there is also good agreement with previous estimates obtained by summing the contributions of individual events. However, the displacement rate in the margin convergence zone of southern British Columbia, Washington and Oregon computed from the seismicity is at least a factor of 10 lower than from plate models and from other convergence estimates, and primarily aseismic slip is suggested. The fault motions as a function of time computed from the seismicity records have also been plotted and compared to the longterm average rates. The plots permit estimates of the minimum present accumulated elastic strain, and show if there are any temporal relations among earthquake displacements on different fault zones. htroduc tionA basic understanding of the origin and nature of earthquakes in a region and detailed evaluation of the associated seismic risk requires development of valid tectonic models. Tectonic

New probabilistic seismic ground motion maps of Canada, displaying peak horizontal acceleration and peak horizontal velocity at a probability of exceedence of 10 percent in 50 years, have been recommended as the replacement for the 1970 Seismic Zoning Map in National Building Code applications. This report presents a comprehensive description of the basic earthquake data and the methods employed in deriving the new maps.

[1] Earthquake hazard in the Cascadia subduction zone forearc comes from three sources: great subduction earthquakes, Wadati-Benioff slab events, and earthquakes in the forearc crust. This study deals with a concentration of forearc crustal earthquakes in the PugetGeorgia region of southwestern Canada and northwestern United States. These earthquakes are due to margin-parallel shortening, rather than compression in the direction of plate convergence. The frequency of large earthquakes has previously been based mainly on extrapolation of the statistics of smaller events from the short instrumental record. In this study, an independent estimate has been obtained through the seismic moment rate required to accommodate current rates of deformation from GPS and geological data. The catalogue statistics to Mx = 7.5 give a moment rate of 4.0 Â 10 17 Nm/yr and a shortening rate of 2.9 mm/yr if all deformation is seismic. GPS data indicate local current N-S shortening of 3 ± 1 mm/yr. For a seismogenic thickness of 12 km, this deformation represents a moment rate of 4.1 Â 10 17 Nm/yr. The predicted occurrences are 0.022/yr (45 years) for M > 6, and 0.0025/yr (400 years) for M > 7, within the uncertainties of the catalogue statistics. Large characteristic earthquakes are not required. Forearc long-term average motion based upon larger-scale GPS, paleomagnetic, and geological data ranges from 5 to7 mm/yr. The estimated long-term occurrence of large events thus is approximately double the current rate, and we infer that the extra earthquakes could follow abrupt N-S shortening in this part of the forearc associated with the oblique rupture motion of great subduction events.

Maximum likelihood estimation of the earthquake parameters No and β in the relation N = No exp (−βm) is extended to the case of events grouped in magnitude with each group observed over individual time periods. Asymptotic forms of the equation for β reduce to the estimators given for different special cases by Aki (1965), Utsu (1965, 1966), and Page (1968). The estimates of β are only approximately chi-square distributed. For sufficiently large numbers of events, they can be estimated from the curvature of the log-likelihood function. Sample calculations for three earthquake source zones in western Canada indicate that for well-constrained data sets, the often-used, least-squares estimation procedures lead to compatible results, but for less well-defined data sets, the effect of subjective plotting and weighting methods used for least-squares fitting leads to appreciably different parameters.

We summarize the methods being used for new seismic hazard maps of Canada, tabulate final values of the 50th and 84th percentile ground motions for major cities, and give uniform hazard spectra, ail for sites on firm soil for both 10% and 2% probabilities of exceedence in 50 years. The availability of strong ground motion relations for spectral parameters allows computation of spectral acceleration maps, which are being recommended as input to the seismic provisions of the National Building Code.