Subnormal Cenozoic geothermal gradients in the extinct Sierra Nevada magmatic arc: Consequences of Laramide and Post‐Laramide shallow‐angle subduction

Abstract

From 120 to 80 Ma, the Sierra Nevada of California was the site of a vigorous, subduction‐generated magmatic arc. At about 80 Ma, arc magmatism left the Sierra and began to migrate far to the east into the interior of the North American plate, where sporadic magmatism occurred during the latest Cretaceous to early Tertiary Laramide orogeny. This migration of magmatism inland has generally been attributed to a shallowing in the angle of descent of the subducting plate. Apatite and sphene fission track thermochronology data indicate that geothermal gradients in the Sierra remained relatively high for 5–10 m.y. after the cessation of Sierran arc magmatism, then decreased to about one‐fifth or less of previous values over about the next 10 m.y. The timing, magnitude, and time scale of this cooling are consistent with conductive cooling of the Sierran lithosphere in response to initiation of unusually shallow underflow of a cold subducting plate below. Dimensional analysis of the equation for conductive heat transfer shows that the thickness of the Sierran block above the shallowly subducting plate would be the dominant factor controlling the expected timing and time scale of subduction‐induced cooling. Modeling of the cooling, supplemented by modern crustal thickness data, yields a best estimate of the Laramide depth to the top of the subducting plate of about 35–50 km, with depths of about 35–60 km permitted. These depths are much less than the likely ≈ 120 km depth of the subducting plate beneath the Sierra during Sierran arc magmatism and thus support the shallow subduction model. Apatite fission track length data allow estimation of paleotemperatures experienced by the Sierran samples in the Tertiary. The estimated paleotemperatures and their variation with sample elevation suggest that even the highest‐elevation samples were buried significant amounts (perhaps 2–3 km) at ≈30–15 Ma and that geothermal gradients at that time were ≈ 5°–15°C/km. Early Tertiary values for gradients were probably similar or lower. The low gradients are an expected consequence of moderately shallow subduction continuing beneath the Sierra through the Cenozoic until the shift to San Andreas transform motion in the late Cenozoic. The evidence for low Cenozoic gradients in the Sierra, in conjunction with evidence for low Late Jurassic to Late Cretaceous gradients in the adjacent Franciscan accretionary prism and Great Valley forearc basin, suggests that gradients in the forearc of the Franciscan subduction zone were continuously subnormal from the initiation of Franciscan subduction in the Late Jurassic until the termination of subduction in the late Cenozoic. This suggests that subduction at the Franciscan trench was sufficiently fast and continuous to maintain continuously low gradients. If gradients remained continuously subnormal, high‐P/T metamorphism in the Franciscan accretionary prism may have occurred throughout the period of subduction and high‐P/T rocks may have resided at depth for extended periods without being overprinted by normal‐P/T conditions.