Abstract Large ice fields (>25 km 2 ) formed over the Tazaghart and Iouzagner plateaus of the High Atlas, Morocco during the Late Pleistocene. The plateau ice fields were drained by large valley glaciers forming a series of moraine assemblages. Four moraine units have been mapped and subdivided on the basis of their morphostratigraphy and the degree of soil weathering. Soil profile development index values indicate that the moraine units are widely separated in time; the oldest moraines are deeply weathered and degraded, whereas soils are absent on the youngest moraines. The highest moraine unit was formed by a small niche glacier that was present as recently as the mid-twentieth century. The Pleistocene glaciers are likely to have been associated with wetter conditions than today and colder air temperatures. Combined with ice in neighbouring areas, such as the Toubkal massif, the SW High Atlas supported some of the largest glaciers in Africa during the Pleistocene. The extent of glaciation, with ice exploiting and breaching drainage divides, has major implications for landscape development. The evolution of the High Atlas has been strongly shaped by glaciation that was closely intertwined with tectonic, fluvial and slope processes.

Abstract Flood events within rivers are responsible for much erosion and deposition. Thus, deposits laid down during floods could potentially comprise the bulk of the Quaternary fluvial record. However, it is difficult to detect individual flood events, as effectively illustrated by the Middle Devensian (Weichselian) to Holocene fluvial sequence from the Nene Valley, Northamptonshire, described in this paper. This is due to limits in the resolution of sedimentological, palaeontological and geochronological techniques. Geochronological techniques have the highest resolution, but error bars of c . 50 years (radiocarbon) and up to 2 ka (optically stimulated luminescence) in the Late-glacial do not allow detection of floods lasting only a few weeks or less. Geochronology is, however, essential for linking periods of fluvial deposition to climatic phases at the marine isotope substage scale. Thus, multiple age determinations show remnant Middle Devensian deposits within a facies association mainly of Younger Dryas age, showing similar fluvial response to climate during both time periods. Palaeontological assemblages suggest that climate was also similar, although with some subtle differences. Determining ‘average’ fluvial activity in response to broad climate phases improves understanding of how rivers behave over long time periods, even though determination of the role of flood events in the Quaternary fluvial record remains elusive.

Abstract The Early-Middle Pleistocene transition ( c. 1.2–0.5 Ma), sometimes known as the ‘mid-Pleistocene revolution’, represents a major episode in Earth history. Low-amplitude 41-ka obliquity-forced climate cycles of the earlier Pleistocene were replaced progressively in the later Pleistocene by high-amplitude 100-ka cycles. These later cycles are indicative of slow ice build-up and subsequent rapid melting, and imply a transition to a strongly non-linear forced climate system. Changes were accompanied by substantially increased global ice volume at 940 ka. These climate transformations, particularly the increasing severity and duration of cold stages, have had a profound effect on the biota and the physical landscape, especially in the northern hemisphere. This review assesses and integrates the marine and terrestrial evidence for change across this transition, based on the literature and especially the following 17 chapters in the present volume. Orbital and non-orbital climate forcing, palaeoceanography, stable isotopes, organic geochemistry, marine micropalaeontology, glacial history, loess-palaeosol sequences, pollen analysis, large and small mammal palaeoecology and stratigraphy, and human evolution and dispersal are all considered, and a series of discrete events is identified from Marine Isotope Stage (MIS) 36 ( c. 1.2 Ma) to MIS 13 ( c. 540–460 Ma). Of these, the cold MIS 22 ( c. 880–870 ka) is perhaps the most profound. However, we here endorse earlier views that on practical grounds the Matuyama-Brunhes palaeomagnetic Chron boundary (mid-point at 773 ka, with an estimated duration of 7 ka) would serve as the best overall guide for establishing the Early-Middle Pleistocene Subseries boundary.