Low heat flux and large variations of lithospheric thickness in the Canadian Shield

Abstract

[1] Ten new heat flux determinations have been made using measurements in 22 mining exploration boreholes located at latitudes higher than 51°N in the Canadian Shield. They provide data in poorly sampled regions near the core of the North American craton where one expects the lithosphere to be thickest. The new heat flux values are all smaller than 34 mW m −2 and are among the lowest recorded so far in the shield. For all the new sites, there is no relationship between heat flux and heat production in surface rocks. In the Canadian Shield, heat flux variations occur at wavelengths <100 km and are mostly of crustal origin. Local averages in two 250 × 250 km windows located on Archean areas at high latitudes on either side of James Bay are 29 mW m −2 and 31 mW m −2 , the lowest values found so far at this scale in the Canadian Shield. S wave traveltime delays derived from tomographic models provide the additional constraints needed to resolve differences of deep lithospheric thermal structure. There is no significant correlation between average surface heat flux and traveltime delays within the Canadian Shield, confirming that variations of the surface heat flux are mostly of crustal origin. Traveltime delays cannot be explained by variations in crustal heat production only and require variations of heat supply to the lithosphere and/or radiogenic heat production in the lithospheric mantle. These variations are associated with changes of lithospheric thickness that may be as large as 80 km. The heat flux at the base of the Superior lithosphere is constrained to be 11 ± 2 mW m −2 .

Temperature and rheological properties of the mantle beneath the North American craton from an analysis of heat flux and seismic data

“…We have assessed the robustness of the solutions in different ways [Lévy et al, 2010]. We did not use the full seismic velocity profiles and preferred to use a single datum such as the traveltime delay instead, because it is more robust to the inversion procedure, as errors within one depth range are compensated by deviations in other depth ranges.…”

“…The choice of this depth interval is justified by Lévy et al [2010]. We expect that traveltime anomalies are restricted to the lithosphere and that no variations exist in the well-mixed convecting mantle.…”

“…This requires determination of average heat flux and heat production values that are representative of the bulk crustal heat production. Lévy et al [2010] have discussed these issues in great detail and have determined averages over 10 geographical windows with the appropriate characteristics. These windows, with typical dimensions of 250 × 250 km, are distributed over the whole Canadian Shield (Figure 1) and contain enough measurements to ensure the smoothing of small-scale crustal heterogeneities.…”

“…We allow the Moho heat flux (Q m ) to vary between 12 and 18 mW m −2 . This range is derived from several independent analyses [e.g., Russell et al, 2001;Mareschal and Jaupart, 2004;Perry et al, 2006b] and is extensively discussed in Appendix C of Lévy et al [2010]. For each set of values of surface and Moho heat fluxes, heat production in the upper crust (A uc ) is readily calculated according to the following equation:…”

“…The method and basic objective are similar to those of Lenardic [1997], Doin et al [1997], Jaupart et al [1998], and Lee et al [2005], but we benefit from a much more extensive geophysical data set as well as a much more comprehensive set of rheological parameters due to Korenaga and Karato [2008]. We account for uncertainties in the large number of parameters involved and use robust scaling laws derived Lévy et al, 2010]. (b) Map of S wave traveltime delays for the depth interval 60-300 km, from the NA04 tomographic model of van der Lee and Frederiksen [2005].…”