THE LATE CENOZOIC UPLIFT AND EROSION HISTORY OF SOUTHERN ENGLAND: PHYSICAL MODELLING OF RIVER TERRACE STAIRCASES

River terrace staircases, dating from ~2Ma, are well-documented along the major Pleistocene river systems of unglaciated parts of southern England. In such a region, with low drainage density and minimal erosion of interfluves on this timescale, rivers form near-equivalent quasi-equilibrium profiles during each climate cycle, such that the net incision between successive levels roughly equates to the contemporaneous uplift of the adjacent interfluve land surface. These fluvial deposits, and related shallow marine sediments dating from ~3.5Ma, are dated using magnetostratigraphic, biostratigraphic, amino-acid, and archaeological techniques. Their altitudes indicate relative stability before ~3Ma, followed by uplift at up to ~0.15mm a-1 during ~3-2Ma. Rates then decreased to a minimum around ~1Ma, before a renewed increase to peak rates of ~0.12mm a-1. The total uplift since the Middle Pliocene is spatially variable, the maximum observed being ~300m. The Middle-Late Pleistocene component increases westward from ~60m near London to ~85m in the Hampshire basin, reaching higher values in south-west England. Physical modelling indicates that the post-Early Pleistocene uplift has been caused by inflow of lower crust, induced by loading by ice sheets in adjacent areas. The required crustal thickening in the London area on this time scale requires an effective viscosity in the lower continental crust of ~1020Pa s, consistent with a Moho temperature just above ~500°C as is reasonable beneath the London Platform. The higher rates farther west reflect the warmer lower crust in these regions, which can flow faster in response to similar lateral pressure gradients. Further physical modelling indicates that the Late Pliocene uplift was much too fast to be sustained by the same process. However, the actual mechanism responsible is more difficult to establish, due to the very limited availability of data from the key interval of ~3-2Ma. This uplift phase may have instead resulted from inflow of lower-crust induced by denudation during ~3-2Ma, when we estimate that a typical layer maybe only tens of metres thick was removed.