Crustal structure of the Celtic Sea and western approaches from gravity data and deep seismic profiles: Constraints on the formation of continental basins

Abstract

A free‐air anomaly map of the Celtic Sea and the Western Approaches is presented including all available marine gravimetric data and Bouguer anomalies on land. N60° features are predominant in the area. They correspond to the N60° so‐called Caledonian grain which played a major role in the Variscan orogeny and in the formation of the Permo‐Triassic sedimentary basins. The fundamental new point which comes out from this gravimetric map is the existence of a closely spaced pattern of N105° features which we regard as late Carboniferous transpressional structures. In the eastern pan of the Celtic Sea, thrusts are in the same N105° direction. West of 6°W longitude, however, folds and thrusts trend preferentially in a N60° direction, following the older grain. There, the N105° lineaments may be regarded as dextral shear structures with a possible compressive component. The Hercynian shortening of the Celtic Sea domain, located between the Variscan Front and the Lizard suture, resulted from the displacement, along the N160° direction of the two adjacent rigid domains. North of the Variscan Front, the N105° and N135° features were possibly active in more recent times. Then, both the N60° and N105° trends controlled the formation of Permo‐Triassic basins, the N105° directions being transfer faults. Two‐dimensional gravity modelling of the SWAT profiles in the Celtic Sea shows that the base of the reflective lower crust may be associated with the Moho discontinuity everywhere in the Celtic Sea except below the Saint George's Channel basin. The lower crust does not change significantly in depth and thickness beneath the Permo‐Triassic basins. None of the available models simply satisfies such conditions. They all require at least a modification of the geometry of the lower crust after the formation of the Celtic Sea basins. Nevertheless, if the simple shear model can be ruled out mainly from structural arguments, the uniform stretching and to a lesser degree the thermomechanical models seem to be more relevant models. However, two additional conditions are necessary: (1) the preservation of crustal seismic features formed before the creation of basins must be explained and (2) a significant portion of the volume of the stretched lower crust (about 40% of this volume for the North Celtic Sea basin) must be added below or perhaps partly above the lower crust by lateral flow of the crust or at the expense of the upper mantle.