Cave levels as proxies for measuring post-orogenic uplift: Evidence from cosmogenic dating of alluvium-filled caves in the French Pyrenees

Abstract

The rates and chronology of valley incision in mountain ranges have been studied in various parts of the globe, but the causes of river incision are often blurred because tectonic, climatic, and sea level-related forcing signals are difficult to distinguish from one another. The Têt River limestone gorge in the Eastern Pyrenees, which displays multiple cave levels containing datable alluvial deposits, provides an opportunity for clarifying this debate. Horizontal epiphreatic passages in limestone can be used as substitutes for fluvial terraces because they correspond to former valley floors and, therefore, also record the position of former local base levels. In the Têt canyon, the passages are filled with quartz-rich sand and gravel sequences that can be dated by 26Al/10Be burial dating. The canyon has cut into a Middle Miocene pediment system—now forming a raised plateau at 1250–1500m—and displays nine cave levels over a vertical height of 1km. One alluvial fill sequence in a cave at +270m above datum (i.e., the local river bed) yielded a weighted mean age of 5.14±0.41Ma; another, situated at +110m above datum, yielded weighted mean ages of 2.23±0.230Ma and 1.20±0.286Ma. The data convert to a mean incision rate of ~52m·Ma−1 since the beginning of the Pliocene, and involved an acceleration to 92m·Ma−1 during the Quaternary. Pre-burial catchment denudation rates range from 35 to 7m·Ma−1, and these also doubled during the early Quaternary. It is concluded that: (i) valley incision into the Miocene pediment has been occurring since 5, probably 10Ma; (ii) there is no evidence of a Messinian canyon in the Villefranche gorge, strongly suggesting through various additional indicators that interference of the Messinian Salinity Crisis with the canyon incision history was minimal; (iii) valley deepening was not a steady process, and recorded periods of stability around 1–2Ma and perhaps 6–5Ma; and (iv) the terraced network of epiphreatic cave levels is primarily explained by tectonic uplift. It follows that the elevated erosion surfaces of the Pyrenees, such as the Miocene pediment directly situated above the canyon edge, were not shaped at high elevations, e.g., by ‘altiplanation’; they formed, instead, close to base level and were uplifted in successive stages by tectonic processes. The study emphasizes the more general proposition that tectonic signals (as opposed to climatic or eustatic) in valley-incision chronologies are best singled out at locations situated among the outer ranges of mountain belts, i.e., in canyons such as the Têt, that respond immediately to base level changes relative to the adjacent foreland. In the inner ranges, fluvial incision is more likely to be affected by the interference of climatic factors (e.g., glaciers), or to be delayed by bedrock impediments to upstream-propagating knickpoints.