The long-wavelength admittance and effective elastic thickness of the Canadian Shield

Abstract

The strength of the cratonic lithosphere has been controversial. On the one hand, many estimates of effective elastic thickness (Te) greatly exceed the crustal thickness, but on the other the great majority of cratonic earthquakes occur in the upper crust. This implies that the seismogenic thickness of cratons is much smaller than Te, whereas in the ocean basins they are approximately the same, leading to suspicions about the large Te estimates. One region where such estimates have been questioned is the Canadian Shield, where glacial isostatic adjustment (GIA) and mantle convection are thought to contribute to the long-wavelength undulations of the topography and gravity. To date these have not been included in models used to estimate Te from topography and gravity which conventionally are based only on loading and flexure. Here we devise a theoretical expression for the free-air (gravity/topography) admittance that includes the effects of GIA and convection as well as flexure and use it to estimate Te over the Canadian Shield. We use wavelet transforms for estimating the observed admittances, after showing that multitaper estimates, which have hitherto been popular for Te studies, have poor resolution at the long wavelengths where GIA and convection predominate, compared to wavelets. Our results suggest that Te over most of the shield exceeds 80 km, with a higher-Te core near the south-west shore of Hudson Bay. This means that the lack of mantle earthquakes in this craton is simply due to its high strength compared to the applied stresses.