Heat production and heat flow in the mantle lithosphere, Slave craton, Canada

Thermobarometric data for mantle xenoliths from a kimberlite pipe in the NWT, Canada are used to constrain the thermal properties of the lithospheric mantle underlying the Slave craton. We derive an analytical expression for a steady-state conductive mantle geotherm that is independent of the geomet...

Full description

Saved in:
Bibliographic Details
Published in:Physics of the earth and planetary interiors Vol. 123; no. 1; pp. 27 - 44
Main Authors: Russell, James K., Dipple, G.M., Kopylova, M.G.
Format: Journal Article
Language:English
Published: Elsevier B.V 01-03-2001
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Thermobarometric data for mantle xenoliths from a kimberlite pipe in the NWT, Canada are used to constrain the thermal properties of the lithospheric mantle underlying the Slave craton. We derive an analytical expression for a steady-state conductive mantle geotherm that is independent of the geometry and thermal properties of the crust. The model has an upper boundary coincident with the MOHO at a depth Z m and has temperature T m and heat flow q m. The mantle is assumed to have constant radiogenic heat production ( A) and we allow for a temperature-dependent thermal conductivity [ K( T)= Ko(1+ B( T− T m))]. Inverting the thermobarometric data through the model geotherm gives limiting values for mantle heat production ( A) and bounds on the temperature dependence of K (e.g. B) that are consistent with the mantle P– T array. We characterize the Slave lithospheric mantle in terms of three critical parameters q m (mW m −2), A (μW m −3), T m (°C). The optimal solution has values [15.1, 0.012, 455]. This characterization of thermal state of the Slave mantle is based mainly on petrological data and is not biased by assumptions about crustal thermal properties. Our analysis shows that a substantial range of parameter values can be used to describe the data accurately and the two bounding solutions are [24.2, 0.088, 296] and [12.3, 0, 534], respectively. However, model parameters are strongly correlated and this precludes the arbitrary selection of values of [ q m, A, T m] from these ranges.
ISSN:0031-9201
1872-7395
DOI:10.1016/S0031-9201(00)00201-6