Subsidence of the Paris Basin and its bearing on the late Variscan lithosphere evolution: a comparison between Plate and Chablis models

Abstract In this study, two different models for the lithospheric thermal evolution are used to predict the long-term subsidence of the Paris Basin. The Plate or Chablis models have a constant temperature or constant heat flow at the base of the lithosphere. These boundary conditions yield a different thermal relaxation time for a given equilibrium lithospheric thickness. The final thermal steady state does not depend on the boundary condition fixed at the base of the lithosphere. The thermal evolution of the lithosphere is computed, taking into account temperature- and pressure-dependent thermal characteristics, metamorphism in the crust, top-crustal erosion, and phase transition in the mantle (spinel–garnet). The long-term tectonic subsidence of the Paris Basin constrains the Mesozoic thermal evolution of the lithosphere and the amplitude of the initial thermal anomaly. The Chablis model, including the effects of spinel–garnet phase transition, predicts present-day lithospheric thickness in better agreement with observation than the Plate model. The Chablis model involves a basal heat transfer of 37 mW/m 2 , similar to the value beneath oceanic plates. The long-term subsidence of the Paris Basin obviously results from the decay of a thermal anomaly initiated during late Variscan times. We first assume that, at 300 Ma, the lithosphere was less than 150 km thick, and the thickened crust started returning to its present thickness (35 km) by uniform lithospheric extension. The subsidence data can be explained by short- (Stephano-Autunian) as well as long (Stephano-Triassic)-lasting extension. These hypotheses both implicitly refer to extensional collapse of the Variscan belt. However, the evolution of the Variscan belt from 350 to 300 Ma may have involved almost no crustal thickening, but an anomalously thin lithosphere (ca 55 km). The corresponding thermal evolution of the lithosphere is computed using the Chablis model with an initially high heat transfer of 61 mW/m 2 at the base of the base of the lithosphere. At 280 Ma, this heat transfer is tuned to 40 mW/m 2 , leading to lithospheric cooling from 280 to 30 Ma, which can also explain the Paris Basin subsidence. For the short-lasting extension hypothesis, the late Variscan lithospheric evolution predicted with the Chablis model yields paleogeography, magmatism, and retrograde P – T – t paths in good agreement with those reported from areas around the Paris Basin. Conversely, the Plate model does not allow mantle melting during the end of Carboniferous and Permian times. The long-lasting extension hypothesis does not explain the paleogeographic evolution, the timing of magmatism, and retrograde P – T paths showing isothermal decompression. The hypothesis with a strong lithospheric delamination cannot be ruled out. However, it is less plausible than the extensional collapse hypothesis.