Lithospheric structure of the Labrador Sea from constrained 3-D gravity inversion

Regional inversions of free air gravity data constrained by bathymetric and sediment thickness information were undertaken over the Labrador Sea and its margins to generate 3-D density anomaly models to investigate broad-scale crustal structural variations across the extinct spreading centre. Benchmarked against independent seismic Moho depth constraints, a density anomaly isosurface within the inverted volumes was selected as a Moho-proxy and regional maps of Moho structure were developed. Inversions using two different sources for depth to basement constraints revealed similar Moho structures with a depth to Moho of 12 km beneath the Labrador Sea which deepens to 20 km and greater towards Davis Strait and beneath the offshore extension of the Grenville Province. Density anomaly slices through the models corresponding to seismic lines show good agreement between the inverted Moho-proxy and the seismic Moho, with the only exceptions occurring where a high velocity lower crustal zone or underplate has been modelled from wide-angle reflection/refraction profiling studies. The inverted depth to Moho estimates were combined with depth to basement constraints to investigate crustal thickness, both for the full crust and for individual crustal layers, revealing that the crust of the Labrador Sea is generally 5-10 km thick but thickens to 20-25 km towards Davis Strait and beneath the offshore extension of the Grenville Province, not taking into account high density underplates or anomalously high density lower crust. Sediment and crustal thickness variations were investigated to compute stretching factors, beta, across the Labrador Sea and to identify zones which deviate from local isostatic compensation. Assuming both an initial unstretched crustal thickness of 35 km and using a variable unstretched crustal thickness model, much of the Labrador Sea has experienced 70-90 per cent thinning. The derived beta values suggest that embrittlement of the entire crust and serpentinization of the upper mantle are likely to have occurred for large portions of the central and southern Labrador Sea, inboard of known oceanic crust. Isostatic considerations reveal sediment excess and deficiency on the Labrador and Greenland margins, respectively, possibly reflecting fundamental rheological asymmetry, with the Greenland margin appearing weaker than the Labrador margin. A strong gradient from sediment deficiency to excess, exclusive to the Labrador margin, may reveal the southwestern limit of a regional graben bounding listric detachment or of a zone of distributed faults and detachments.