Tertiary Rivers: Synthesis
Following the present review of evidence derived from Tertiary sedimentation patterns, it seems advisable to adjust the perspectives provided by the more exclusive reliance on morphological evidence in a number of significant respects. Viewing drainage and land surface evolution as a component of a coupled erosion and depositional system, it is to be expected that the nature of erosional activity will be reflected in depobasins which will also contain evidence of terrestrial environments (such as the nature of landmass-yielded sediment and alluvial sedimentation style) that is simply not available from analysis of fragmented erosion surfaces and their in situ residual deposits. This is amply illustrated in the considerable body of geological research which has been reviewed here.
The Wooldridge and Linton model has already been much modified. In particular, erosional surfaces are now more generally regarded as composite and in part of Palaeogene origin. Jones’s recent (1999a) synthesis identifies a series of events in a “new model of Tertiary landscape evolution”. This begins with emergence of the land from beneath the sea that began during the late Cretaceous and was completed by the early Palaeocene. This was a dominating phase that provided the land surface upon which the drainage system developed. Destruction of the extensive Chalk up to 350 m thick followed, with virtually the entire cover being removed over uplifting areas (e.g. the Weald-Artois anticline, the Channel High), under a tropical to sub-tropical climate. Palaeocene and Eocene sediments were deposited on a polygenetic sub-Palaeogene surface. By this time, the surface had a subdued relief form overlain by a weathered regolith. Jones also emphasises the identity of separate morphotectonic regions differentiated to the extent that regional uniformity, specifically in chalkland areas, is unlikely.
The present synthesis emphasises the persistence of coupled erosion and sedimentation during a whole series of transgressive and regressive episodes in the Palaeogene rather than the separation of erosion and sedimentation periods: it appears doubtful whether periods of uplift/ high relief were separated from ones of erosion/low relief in but a few stages echoing the Davisian cycle of erosion, even in modified form. Tectonic deformation occurred in pulses throughout the Palaeogene, with structural basins (e.g. the London and Hampshire-Dieppe Basins) becoming more strongly defined by the growth of the Weald-Artois anticline and the Isle of Wight monocline. Sedimentation was focused on these basins, deposition in the London Basin ending in the Eocene, presumably determined by available accommodation, but continuing into the Oligocene in the Hampshire-Dieppe Basin. There was an extensive, and probably facetted, low-relief duricrusted land surface, which in the west was episodically denuded from the Palaeocene onwards, but which in the east developed on younger emergent Palaeogene sediments. But formerly postulated periods of erosion (as in the production of a mid-Tertiary peneplain and a later marine transgression) without evidence of substantial deposition appear unsupportable and they are not reflected in the actual sedimentary record as strongly as are many other episodes.
Apart from the wealth of information now available on Tertiary rivers and their environments, perhaps the single most striking point to arise from this synthesis is the long-term stability of the fluvial system, the major elements of which were clearly already in existence in the late Palaeocene. They must therefore have been established on the emerging and deforming surface at the end of the Cretaceous. The present assessment of river development from the standpoint of fluvial sedimentation and provenance, rather than surface morphology, unequivocally demonstrates that the major elements, the Thames, Solent, Irish Sea river and possibly an early Trent river existed throughout the Tertiary, i.e. for at least 55 million years and indeed on into the Pleistocene. This conclusion must be seen against a back-drop of significant, continual crustal deformation and upwarping throughout the period that continues today. It is even more notable that the Solent River has entered the Channel area in virtually the same place at periods of active erosion / sedimentation throughout this period. The Thames has, by contrast, markedly extended and reduced its course in response to external changes; a point discussed in a Pleistocene context.
Tectonic regime has been the overriding control on the long-term stability of the drainage system. Notwithstanding the major changes that have taken place, the region has effectively remained in the same tectonic setting since the end of the Cretaceous, i.e. the opening of the North Atlantic Ocean basin and during the later Alpine (Helvetic) Orogeny. The interaction of these processes has produced the long-term uplift of north-western block areas, the rejuvenation of Variscan structures south of the Variscan Front (from south Wales to central Kent, southern England) and the long-term continued downwarping of the North Sea. The net effect has been to cause southern Britain to tilt towards the south-east throughout the Cenozoic. It has also led to repeated movement on some critical more localised structures. Apart from the inversion of a series of basins noted above, it has driven the continual, pulsed rise of the Weald-Artois Anticline, a major structural element that has influenced the palaeogeography of south-eastern England and in particular the drainage evolution and the seaway form virtually throughout the Tertiary. This structure began rising in the Early Eocene Ypresian Stage, becoming a significant barrier in the Eocene Middle Lutetian, following the London Clay sea-level highstand, and remaining so until it was breached in the Middle Pleistocene. The barrier has conditioned the separation of the Solent and Thames drainage for 50 million years. Moreover, drainage developed concordantly to the macrostructures and regional slopes of the elongated dome-like ridge radiating predominantly north and southwards to enter the precursors of the North Sea and the Channel with probable antecedence operating with effect to minor folds.
Similarly the substantial Wight-Bray Monocline forms part of a major series of compressional structures that reflect reaction of deep-seated Variscan crustal structures. The predominantly upward movement on this west-east, then NW-SE-trending feature has controlled the southern margin of the Solent system and of the Hampshire-Dieppe Basin throughout the Cenozoic.
Long-term consistency is seen not only as continuity of course-alignment but also of river form. Surprisingly, meandering streams predominate throughout the period and the evidence conforms generally to models for rivers in seasonal tropical / sub-tropical/ warm-temperate situations. Although their form will inevitably have varied locally depending on materials in transport, discharge variability and variations in slope, the overall stability in fluvial form is striking across the region. This stability appears to relate to the subdued relief that seems to have prevailed, to stability of regolith resulting from dense vegetation cover, and to predominance of fine particulate materials in transport. It is clear that during certain periods, e.g. the late Ypresian to Lutetian or the middle Thanetian, the rivers were predominantly transporting clastics, sand and silt derived from destruction of the uplifting hinterland to the west, particularly Palaeozoic and Pre-Cambrian massifs, resulting in great expansions of sand-rich deltas. By contrast during the early Ypresian, the dominant load was clay, derived from the thick weathering crusts, and deposited as London Clay when a period of comparative structural quiescence coincided with the maximum of Palaeogene marine transgression. Similarly, there also seem to have been periods, such as during deposition of the Bouldnor Formation (Rupelian) of the Isle of Wight (Fig. 3), when only very limited volumes of fines were in transport. Where channel and floodplain complex depositional sequences have been recorded, e.g. the Reading Beds (late Thanetian), a mixed load of fine granular material, together with sands, fines and locally clay-breccias was being moved. In almost all the cases described, vegetation played a profoundly important role. Repeatedly descriptions of lignite (peat), which may be associated with floodplain sequences, are reported. These sediments can arise in various ways, most commonly by in situ growth and accumulation in floodplain pools or hollows where they frequently represent the end-member in fining-upward cycles, e.g. in the Bovey Basin Oligocene sequences. However they can also originate as transported, detrital organic accumulations, such as those described by from Wareham, or from Felpham. In most cases the dense vegetation can be assumed to have greatly contributed to bank cohesion and stability; a situation typically found in tropical, sub-tropical and warm-temperate environments alike.
Floodplain surfaces are also repeatedly represented by descriptions of palaeosols, mostly originating under moist, but seasonally dry conditions, e.g. in the Hampshire Basin Reading Beds (Palaeocene), the Headon Hill Formation (late Middle to Late Eocene), in the Bovey Formation (Oligocene) and in the Lignite and Clay Unit (Oligocene – Early Miocene) in the Cardigan Bay Basin. Lateritic soil processes and silcrete formation, operating under a savannah (semi-arid) climate with intermittent dry periods and perhaps areally restricted along drainage lines, have also been reported from the early Middle Eocene of Devon, whilst duricrust remnants involving a variety of parent bedrock materials are well known.
Coarser accumulations, dominated by gravels (conglomerates), are extremely rare in the regional Tertiary alluvial record. It is only in the braided sheet-flood sequences of Devon, the pebbly sands of the Reading Formation and the fluvial fan-type sequences in proximity to active faults, such as the Mochras Fault in Cardigan Bay, the Bracklesham Group in Dorset and the Sticklepath –Lustleigh Fault in Devon, that coarse aggradations apparently occurred. Whilst this may be an artefact of their very low preservation potential, because such fan-type sequences are usually restricted to areas of high relief which are latter removed, it is more likely the result of general and persistent subdued topography simply lacking sufficient potential energy to generate flows capable of moving coarse detritus. This suggests that even under conditions during which precipitation was far higher than that today , storm-induced floods were seldom able to cause substantial movement of gravel-sized material. Also to be emphasised was the density of the vegetation cover throughout that acted to cushion flooding by removing water by evaporation, supported by efficient groundwater percolation. In addition, the predominance of intense chemical weathering through the period may have resulted in coarse gravel-sized material being relatively rare in the landscape. Where coarse clastic materials do occur they overwhelmingly comprise chemically-stable lithologies, except in particular local situations, such as the alluvial fans of north Wales.
As noted above, the Tertiary deposits of lowland Britain, particularly in the main depositional Hampshire-Dieppe and London basins, characteristically record alternating transgressions and regressions that have been attributed to global eustatic sea-level cycles, modified by local tectonic activity. The fluvial responses to these sea-level changes appear to parallel closely the reactions seen in British Pleistocene river systems, i.e. during transgressions their lower valleys are drowned by the sea, and during regressions the rivers extend their courses across the highstand sediments, in places cutting through the pre-existing marine/estuarine/deltaic sediment wedge en route, as they establish a graded course to the sea. Evidently, the rivers did this repeatedly throughout the period.
The general observation that, unlike nowadays, low-relief land surfaces dominated the southern British region until the Pleistocene is extremely important. Today most of the area, although not mountainous, has a considerable topography. Given that the tectonic regime currently affecting the region has remained broadly the same throughout the Tertiary, it is apparent that today’s deeply-incised river valleys must be the product of high, predominantly coarse to very coarse sediment yields, resulting from the substantial, rapid climate changes that characterise the Pleistocene. These climates introduced permafrost and cold-climate weathering products to river systems. The frequent thinning and occasional disappearance of the vegetation cover and the altered conditions for channel-bed and valley-floor incision that accompanied these changes explains the landscape dissection. Although without doubt substantial climate change also occurred in the Tertiary, e.g. in the Oligocene, the changes apparently had less impact on long-term landscape evolution. This highlights the significance of mechanical (nivation) weathering compared with chemical weathering for the rate of landscape dissection and lowering. Moreover, it emphasises the role of forest vegetation in stabilising the landscape surface whilst reducing surface runoff, even though precipitation was apparently greater throughout much of the Tertiary than during the Quaternary.