Chapter 7
The North Sea

Kim M. Cohen,^1,2,4 Kieran Westley,³ Gilles Erkens,^1,4 Marc P. Hijma⁴ and Henk J.T. Weerts⁵

¹ Utrecht University, Utrecht, The Netherlands

² TNO Geological Survey of the Netherlands, Utrecht, The Netherlands

³ School of Geography and Environmental Sciences, Ulster University, Coleraine, Northern Ireland, UK

⁴ Deltares Research Institute, Utrecht, The Netherlands

⁵ Cultural Heritage Agency, Amersfoort, The Netherlands

Introduction

The southern and central North Sea basins (Fig. 7.1) represent some of the most important areas globally both for the discipline of submerged prehistory and the wider study of the Paleolithic and Mesolithic (Peeters et al. 2009). From their margins come the earliest (as yet) evidence of the hominin occupation of northern Europe in the form of the East Anglian sites of Pakefield and Happisburgh. Both sites are part of the Cromerian Complex stage of the early Middle Pleistocene, and indicate hominin occupation potentially as early as 900 ka to 800 ka, in a coastal plain setting (Parfitt et al. 2005; 2010). From this stage onwards, evidence from other British and European sites (including coastal plain sites e.g. Clacton) also shows occupation on a semi-continuous basis with an increase in intensity from Marine Isotope Stage (MIS) 13 (ca. 500 ka) onwards (Pettitt & White 2012). Even so, throughout the Pleistocene the pattern remains one of repeated pulses of (re)colonization and abandonment and with population shifts involving multiple hominin species including Homo sapiens, Homo neanderthalensis, Homo heidelbergensis and the as-yet unidentified occupants of Pakefield and Happisburgh, possibly H. antecessor (Stringer 2006; Ashton et al. 2014; Roebroeks 2014).

Figure 7.1 Map of the North Sea, showing general bathymetry and modern political geography. Key seabed toponyms are indicated by letters, key archaeological localities by numbers. The bathymetry has been derived from the EMODnet Hydrography portal: www.emodnet-hydrography.eu. Reproduced with permission.

Strongly implicated in this are the former coastal plains of the North Sea, serving not only as a migration pathway into and out of the British Isles at various stages in the Pleistocene (Paleolithic; Parfitt et al. 2010; Cohen et al. 2012; Roebroeks 2014), but as a vast now-submerged lowland offering its own attractions, and, in the Early Holocene, potentially supporting its own major populations (Mesolithic Doggerland; Coles 1998; Peeters et al. 2009; 2014). Crucially, there is strong evidence that the present seabed and sub-seabed within the North Sea contain a rich and well-preserved archaeological record as well as extensive remains of the former associated paleolandscapes and paleoenvironments. Abundant lithic and organic implements, faunal and hominin remains, and peat deposits have been trawled or dredged from the southern North Sea over the past decades. Among the finds are tonnes of Pleistocene and Holocene bones found offshore of the Netherlands, mainly faunal but also including hominin remains (Mol et al. 2006; Hublin et al. 2009), assemblages of Middle Paleolithic handaxes from off East Anglia (Tizzard et al. 2011; 2014), and an in situ Mesolithic assemblage from ca. 20 m water depth in Rotterdam harbor in an area that was just offshore before recent harbor expansion (Weerts et al. 2012; Borst et al. 2014; Moree & Sier 2014; 2015; Vos et al. 2015).

The North Sea is one of the best-studied shelves in the world from a geological perspective. Research triggered by the presence of natural resources such as hydrocarbons and aggregates, and, more recently, by the development of offshore wind farms has generated a vast amount of geoscience data. It is the subject of numerous publications that are relevant to the study of submerged prehistoric landscapes, their Quaternary evolution and their taphonomy. The last decade in particular has seen an increased number of publications dealing directly with paleolandscapes. The work of Gaffney et al. (2007; 2009) demonstrated the use of reprocessed merged industrial 3D-seismics in surveying large swaths of the central North Sea, focusing on the Late Glacial and Early Holocene. The multinational North Sea Prehistory Research and Management Framework (Peeters et al. 2009) inspired various collaborative efforts, including a geological background paper for hominin occupation of the southern North Sea (Hijma et al. 2012). In 2014, a special issue of the Netherlands Journal of Geoscience was devoted to the subject (Peeters & Cohen 2014). New and groundbreaking studies are also underway. In January 2015, the annual discussion meeting of the Quaternary Research Association focused exclusively on the Quaternary of the North Sea, treating a mixed academic and industry audience to many geological and archaeological highlights. In particular, research funded by the Forewind consortium in preparation for a 7 GW wind farm on Dogger Bank will bring our geological and archaeological understanding of the central North Sea to a new level (Forewind Consortium: www.forewind.co.uk/).

Harmonized research into the North Sea as a contiguous sea basin is complicated by its partition into sectors governed by different political entities. Figure 7.1 shows not only the generalized bathymetry, but also the division of the North Sea shelf among seven European countries (France, Belgium, the Netherlands, Germany, Denmark, Norway and the UK). The German part of the continental shelf within the 12 nautical mile zone does not fall under the federal government but under the individual states, adding three more authorities: Lower Saxony, Hamburg and Schleswig-Holstein. Earth science data (high-resolution bathymetry, core logs, 2D and 3D seismics) are at present available in vast numbers (Figs. 7.2 & 7.3) but are highly diverse, collected using multiple techniques and stored in different databases, and owned and maintained by different institutions in each of the countries. In addition, the various data sets have been gathered by workers from many different research traditions. Finally, data is stored not only in national and institutional databases but in less accessible databases held by the commercial users of the North Sea (e.g. oil, renewables and aggregates industries) and their specialist contractors. At the time of writing, various national and international initiatives include efforts to collate and harmonize data and data products from the North Sea, for example EMODnet (European Marine Observation and Data Network 2014). Many international initiatives focus on key national-level data sources (e.g. those managed by national geological surveys). At national level, initiatives try to unlock larger sets of legacy data, including that held by commercial parties (offshore engineering, aggregate mining) for future public use, for example, the UK Crown Estate Marine Data Exchange for offshore renewables (www.marinedataexchange.co.uk/).

Distilling taphonomic information from such a diverse set of databases is a challenge. This chapter offers a general summary of the Quaternary history and geoarchaeological taphonomy of the southern and central North Sea continental shelf, as a starting point to understanding the taphonomy of submerged archaeology from paleolandscapes preserved below the North Sea floor. For further information, the interested reader is directed to the references herein, which provide the underlying information in more detail than is possible here. One entry point is the Audit of Current State of Knowledge of Submerged Palaeolandscapes and Sites, carried out by Wessex Archaeology (Bicket 2013) on behalf of English Heritage, for the UK sector. The aforementioned integrative articles provide further entry points (Gaffney et al. 2007; Hijma et al. 2012; papers in Peeters & Cohen 2014) while earlier review papers or volumes include Flemming (2002; 2004). For nearshore and estuarine processes around the North Sea, an entry point is Martinius and van den Berg (2011). For sea-level research, and related matters such as mapping and dating the drowning of landscapes due to sea-level rise, a handbook edited by Shennan et al. (2015) is a good entry point.

Physical Geography and Modern Environment

The present-day North Sea is a semi-enclosed epi-continental shelf sea occupying an area of ca. 750,000 km².Water depth is generally shallow and deepens as one moves north and north-east (Fig. 7.1). The shallowest area — the Southern Bight, located between southeast England and the Low Countries — is less than ca. 40 m deep. North of this, the central North Sea covering the zone between England and Denmark is characterized by depths of ca. 40 m to 100 m. These areas are shallow seas owing to sustained delivery of sediment over millions of years, outpacing subsidence of the North Sea as a tectonic basin. This filled the basin gradually from the south-east to the north-west, with glaciations modifying the surface in the last million years (see sections below, pages 152–161).

The deepest areas of the shelf are located between Scotland and Norway; water depths here are between ca. 100 m to 200 m and there is a major bathymetric depression in the form of the Norwegian Channel which reaches maximum depths of ca. 700 m (e.g. Huthnance 1991; Paramor et al. 2009). This trench is a continuation of the Oslo fjord and was the path of a major ice stream in the Last Glacial and at least four earlier glaciations. Its depth is inherited from ice-age subglacial erosion (Sejrup et al. 2003).

Generalized (low resolution) bathymetric data indicates a relatively featureless seabed, typified by large expanses of low gradient terrain. The exceptions are upstanding features which range in size and scale from the numerous long and narrow sand banks of the Southern Bight to the plateau-like expanse of the Dogger Bank in the central North Sea. Some bathymetric depressions are also evident on the shelf, most obviously the Outer Silver Pit, a west–east oriented valley incised to depths of ca. 80 m. Smaller depressions, such as the Silver Pit, Sole Pit and Well Hole, are also present (Briggs et al. 2007).

The North Sea is more sheltered than other parts of the European shelf owing to the surrounding landmasses. In general, the seabed substrate varies depending on the hydrodynamic conditions described above in conjunction with bathymetry, inherited geomorphology and substrate type. Based on broad-scale mapping, most of the North Sea floor is characterized by sand. Widespread occurrence of coarser sediments is limited to the eastern English coast. Patchy areas with abundant coarse material also occur off Denmark, Germany and the Netherlands where glacial sediment has been winnowed by marine processes. Patchy gravelly areas also occur in the very south off Belgium (Flanders) and southeast England (Kent) in the entrance of the Dover Strait, due to marine winnowing of fluvial deposits. Fine-grained muds or muddy sands occur over other parts of the basin. Past mapping of the seabed substrate and collated maps based on these (e.g. EMODnet) suffer from the crude resolution of the data which was available at the time. Present-day surveys typically allow major improvements in resolution, and can identify finer-grained patches within coarser sand seas and sandy patches within muddy areas, thus helping to resolve inherited landforms (e.g. Tizzard et al. 2014).

Tides are the dominant control in shelf hydrodynamics and proceed anticlockwise rotating around amphidromic points in the Southern Bight, west of Denmark and at the southern tip of Norway (Otto et al. 1990). Consequently, the timing of tidal cycles varies across the basin, as does the tidal range, which is larger on the British coast compared to continental Europe (Otto et al. 1990; Huthnance 1991). Tidal currents vary correspondingly and are therefore greater along the western side of the sea basin (Huthnance 1991; Neill et al. 2010). Despite the dominance of tidal forces, storms and wind-driven currents also play a part (van der Molen 2002). Dominant wind directions are from the west and south with wind speeds highest in the northern part of the North Sea and decreasing south and east (Neill et al. 2010). When combined with bathymetry, these forces set up bed shear stresses which are greatest in the shallow nearshore regions of the southern North Sea (i.e. off East Anglia, Belgium, the Netherlands, Germany and southern Denmark) (Neill et al. 2010). This is of relevance to paleolandscapes and taphonomy because it controls the height and trough depth of bedform fields (‘sand waves’) which have become established over the last 6000 years (van der Molen 2002). The reworking depth of these sand-wave fields affects preservation of Late Glacial and Early Holocene land surfaces especially where they are not capped by compacted muds or peat.

Spatial variability in inherited topography, geology, past sea-level change, glaciation and modern oceanographic conditions results in significant natural coastal variation which has, more recently, been amplified by human influence. The bounding northern shorelines (Scotland and Norway) are steep, often rocky, with fjords and large inlets (Barne et al. 1997). The Norwegian coast is particularly rugged and complex, indented with numerous bays, inlets and fjords, and dotted with many small islands. Further south, the coastlines become lower gradient and softer and are no longer rocky. The availability of sediment is the result of prolonged supply from the large northwest European rivers and the modification of these rivers by glaciations.

Proceeding in an anticlockwise direction, the coastline of northeast England and the northern half of East Anglia comprises relatively linear unconsolidated cliffs (bluffs) of glacial sediment which, at present, are eroding (Barne et al. 1995a,b). It is in fact this cliff retreat which allowed discovery of the Pakefield and Happisburgh sites (Stringer 2006). The southern half of East Anglia and southeast England are characterized by a more indented shoreline dominated by large estuaries (e.g. Thames, Medway) and their associated salt marshes (Barne et al. 1998; Allen 2000). Many of these estuaries represent drowned valleys flooded by rising sea level after the end of the Last Glacial.

The opposing side of the Southern Bight is characterized by extensive linear sand beaches, commonly backed by dunes. Major drowned valleys existed in this area too (e.g. the Rhine-Meuse valley of Busschers et al. 2007; Hijma & Cohen 2011; the Central Netherlands Rhine valley of Peeters et al. 2015a), beside newly-formed transgressive tidal inlet systems (Rieu et al. 2005; Hijma et al. 2010). However, the matured Holocene barrier system means that from ca. 4500 years ago (Beets & van der Spek 2000), these no longer show up as interruptions in the coastline. Major breaks in this shoreline, which encompasses Belgium and the Netherlands, occur in the southwest Netherlands as a series of estuaries of the Scheldt, Meuse and Rhine (e.g. van den Berg et al. 1996; Martinius & van den Berg 2011). The large number of estuaries here originates from Medieval storm surge incursions that repeatedly caused loss of cultivated coastal peatland (e.g. Vos & van Heeringen 1997, Vos 2015).

Moving north and east along the Netherlands coast through north Germany and into southern Denmark, the coast is marked by barrier islands (the Frisian Islands) which protect the muddy back-barrier lagoon of the Waddenzee. The islands are associated with tidal deltas and tidal inlets, and the Wadden environment with salt marshes, intertidal sandy shoals and tidal mudflats (e.g. Oost 1995; van der Spek 1996; Allen 2000; Behre 2004; Streif 2004; Chang et al. 2006; Vos 2015). A break in the island chain occurs where the Weser and Elbe exit into the North Sea; these too occupy drowned Pleistocene valleys (e.g. Streif 2004; Alappat et al. 2010). Continuing around the coast, western Jutland is typified by mainland sand beaches, lagoons and barriers, with the latter becoming increasingly attached to the mainland as one moves north (Pedersen et al. 2009; Fruergaard et al. 2015), and farthest north by cliff sections exposing landforms from the Last Glacial with a sandy seabed in front (Anthony & Leth 2002).

Quaternary Background and Paleogeographic Framework

Early Pleistocene

The present configuration of the southern North Sea basin is a relatively recent phenomenon. This section outlines the broad landscape and paleogeographical changes taking place in the southern North Sea region over the past ca. 1 million years, coincident with the first hominin occupation, presently confirmed as dating to the Early-Middle Pleistocene transition (Fig. 7.4). Throughout this period, the fluctuating Quaternary climate, sea level and extent of glaciation together with regional tectonics reshaped the landscape on multiple occasions and progressively over multiple glacial-interglacial cycles. From ca. 800 ka onwards the length and intensity of glacial periods increased with attendant effects on ice-sheet extent and glacio-eustatic sea level (Zachos et al. 2001). While preceding ocean volume changes are estimated to have been less than 70 m, from the Middle Pleistocene onward, glacial maxima saw glacio-eustatic lowstands as deep as 120 m to 150 m (Rohling et al. 2009). These figures broadly apply to the North Sea too, although the gravitational effects of the nearby ice masses meant that the regional lowstands were some 10 m to 20 m less deep than the global average.

The climatic changes since the onset of the Middle Pleistocene have controlled not only the paleogeography of the study area, but also the preservation of the sedimentary deposits (and attendant archaeological evidence contained within, i.e. taphonomy) that provide evidence of past landscapes. The remainder of this section will outline the links that exist between geological controls and taphonomy. This prepares for the Submerged Landscape and Preservation/Taphonomy dedicated sections in the second half of the chapter (pages 160–175).

A first major control on long-term geological preservation and paleogeographic change is the tectonic setting. Much of the North Sea area is part of a subsiding basin, initiated by reactivation of a Triassic-Jurassic rift system early in the Miocene (25–20 Ma) (see overview in Cloetingh et al. 2005). The basin is aligned SE–NW running from onshore (the Netherlands, northwest Germany, southern Denmark) to offshore, reaching as far north as southern Norway (Fig. 7.5). No major changes in the tectonic regime have taken place during the Quaternary, with The Netherlands' onshore depocenters subsiding at rates averaging up to 0.3 m/kyr (Kooi et al. 1998), increasing to 0.6 m/kyr in the central North Sea offshore.

Subsidence allowed the accumulation of Quaternary sediments up to several hundred meters thick within the depocenter (Cameron et al. 1992; Gatliff et al. 1994). By contrast, mild uplift predominated around the ‘shoulders’ of the basin: East Anglia, Flanders and the parts of the North Sea between them, as it also did for Mesozoic uplands of the Weald-Artois anticline (e.g. van Vliet-Lanoë et al. 2000; 2002; Cloetingh et al. 2005). This affected paleogeographical evolution and affects paleolandscape taphonomy in the southern North Sea (Cohen et al. 2012; Hijma et al. 2012). The sediment stored in the North Sea depocenter has mainly been deposited by the major northwest European rivers (e.g. Rhine, Thames) and the former Eridanos River (Overeem et al. 2001). This latter ancient watercourse used to drain a catchment spanning the Baltic and Fennoscandia (Fig. 7.6). In doing so, it supplied immense quantities of sediment to the North Sea enabling considerable delta plain progradation. Some 80% of the total accumulation is from this river system (Cameron et al. 1992). The end result was the formation of a wide fluvio-deltaic coastal plain that, at highstand, extended some 100 km northward from the Weald-Artois anticline, and up to 300 km at lowstand. As such, these lowlands were a considerable part of the broad land bridge that connected Britain and continental Europe even during highstands (Funnell 1995; Hijma et al. 2012).

From about 1 Ma onwards, enlarged Fennoscandian ice masses began destroying the Eridanos system and, in tandem, important highstand coastline reconfiguration commenced in the North Sea area (Gibbard 1995; Cohen et al. 2014). These events mark the Early-Middle Pleistocene transition in the study area, and provide a second key control on the lowland environments around the North Sea.

Middle Pleistocene

By 1 million years ago, the chalk hills across the modern Dover Strait still formed a distinct southern boundary to the study area (Figs. 7.6 and 7.7a). This was the time when hominins first reached the North Sea, during the interglacial highstands of the Cromerian Complex (MIS 21–13; Fig. 7.4). Loss of the Eridanos sediment supply at that time meant that the fluvio-deltaic plain was moribund and its surface subsiding (Cameron et al. 1992). The trend of coastal progradation had ceased and highstand shorelines were shifting ever closer towards their present position (Zagwijn 1979; Funnell 1996). Farther south, between the chalk hills and the North Sea highstand shoreline, a considerable stretch of terrestrial lowland habitat was present, linking southern Britain and the European mainland (Funnell 1995; Gibbard 1995). Importantly, marine incursions via the English Channel to the south did not breach the Dover Strait area at this time (Fig. 7.7a).

The highstand shoreline along the southern rim of the North Sea, running from the Netherlands to East Anglia, can be seen as a continuous deltaic coastal plain supported by the extensions of the Rhine, Meuse, Scheldt, Medway and Thames rivers of the time (Cohen et al. 2012). The early sites of Pakefield and Happisburgh were located on the western edge of this coastal plain. Paleoecological data indicate that the environment during (part of) the earliest occupation (ca. 900–800 ka; Happisburgh) was dominated by boreal forest (e.g. pine, spruce) and modestly cooler than these localities are today. Late phases of occupation within the Cromerian Complex stage (Pakefield), took place during periods of interglacial climax, in areas characterized by marsh, oak woodland and open grassland, modestly warmer than these locations are today (Parfitt et al. 2005; 2010). As the Cromerian Complex was the time of earliest demonstrable hominin occupation of the region (Lower Paleolithic; Fig. 7.4), and as sites from that period are encountered in lowland deposits and paleoenvironments (Parfitt et al. 2010), it is paleoanthropologically relevant to generate information on past distribution and present preservation of such deposits in the nearshore and offshore in the modern North Sea (Peeters et al. 2009).

The state of preservation/taphonomy of deposits in coastal plain environments, is controlled by various processes, both syn-sedimentary (during the interglacial concerned) and post-sedimentary (in later glacials and interglacials). With the increased amplitude of glaciations during the Cromerian Complex, the proximity of the North Sea to centers of glaciation increased glacio-hydro-isostatic adjustment (GIA) effects. This had a sustaining effect on relative sea-level rise into highstands, causing transgressions in the North Sea to commence relatively late in the interglacial.

While post-glacial transgressions were initially fast, they subsequently slowed down and continued for a long period. This meant that each interglacial following a large glaciation saw its highstand coastal rims characterized by extended periods of sediment aggradation. This, in turn, produced coastal plain architectures with favorable properties for enclosing archaeological sites (Cohen et al. 2012). As the furthest inland rims of the coastal plain are the last to erode, even from late in the interglacial, sites located there would still have had a good chance of burial by continued coastal aggradation. This enhances the long-term preservation potential for sites from the second half of the interglacial in the North Sea region (whereas in the tropics and subtropics, where GIA differs, this effect unfortunately does not exist, Cohen et al. 2012). In the southern North Sea, GIA effects thus caused late transgressions that gave hominins dwelling in the coastal plain enough time to reach this far north (see also Late Glacial and Holocene, pages 161–164). Furthermore, GIA effects helped to ensure that sites from the second half of the interglacial would still be preserved and located relatively far inland.

The depths of preservation of former coastal plain landscapes in the long run are controlled by tectonic subsidence regimes. In regions lacking tectonic subsidence, highstand coastal plain deposits from past interglacials underwent coastal erosive attack in younger interglacials each time a new highstand North Sea was established. Therefore, around the North Sea the problem of preserving coastal plain archaeology does not lie in its being preserved during the ca. 10 kyr of an interglacial, but its longer-term preservation thereafter (Cohen et al. 2012). In that respect, the role of the Anglian glaciation in preserving Cromerian Complex coastal plain deposits appears to be essential. It is the water-laid nature of these glacial deposits that explains why the tills buried rather than incorporated the older deposits in this particular region, outside the preservation-favourable tectonic area (Fig. 7.5). This relatively favorable preservation control applies to areas offshore East Anglia, once covered by Anglian till, and is a taphonomic reason to pay special attention to the proglacial situation of the Anglian glaciation. A further reason (Gibbard 1988) is that besides controlling the preservation of Cromerian Complex deposits, the Anglian glaciation affected the drainage networks of many of the rivers around the southern North Sea (see next pages).

Cold stages are apparent for the pre-MIS 12 Cromerian Complex, with glaciation likely encroaching on the northern margin of the North Sea basin (e.g. MIS 16) (Rose 2009; Böse et al. 2012). However, the first major verifiable expansion of lowland ice across the study area occurred during MIS 12 (Fig. 7.4; Pawley et al. 2008; Lee et al. 2012). On at least three occasions in the last 500,000 years, ice sheets originating in Britain and Scandinavia extended across the sub-aerially exposed landscape of the North Sea basin. These are the well-documented Anglian/Elsterian, Saalian (at maximum glaciation called the Drenthe substage) and Weichselian limits across the North Sea (Laban 1995; Praeg 2003; Carr et al. 2006; Busschers et al. 2008; Sejrup et al. 2009; Lee et al. 2012; Moreau et al. 2012). Glaciation was also substantial in MIS 10 (Elsterian or earliest Saalian) and MIS 8 (early Saalian), with British ice covering northern England and Scandinavian ice covering Denmark and reaching into the central North Sea and the German Bight, but leaving the southern North Sea ice-free.

The impact of glaciation on the landscape was considerable, not only in ice-covered areas but also in the foreland that received abundant glacial outwash, and was marked by severe periglacial frost conditions mobilizing the surficial sediments on slopes and in rivers. Prior to the large glaciations, many river systems, like the Thames, Rhine, Elbe, Meuse, Scheldt and (now extinct) northern English rivers Ancaster, Bytham and proto-Trent, had drained northwards into the North Sea, reaching the Eridanos Delta lowland (Gibbard 1988; Rose 1994; Funnell 1996; Rose et al. 2002; Cohen et al. 2014). Each large glaciation forced many of the river courses around the North Sea into new, more southerly positions, temporarily or permanent. Valley floors of abandoned courses became used by underfit smaller streams (e.g. Gibbard et al. 2013; Peeters et al. 2015a), and many of them were eventually buried by periglacial colluvium, glacial diamictons (tills) or marine deposits. By governing abandonment and burial of valleys and contemporaneous creation of new pathways, shifting of rivers played a major double role in recording and preserving the archaeology from the various Paleolithic periods (Fig. 7.4), especially in the southern North Sea (Hijma et al. 2012).

Areas directly under the ice were remodeled on each occasion and imprinted with a variety of glacial features including both depositional elements, such as till sheets, and erosional landforms, such as tunnel valleys (see ‘Evidence of submerged landscapes on the shelf’, pages 166–169). Former ice margins are marked by features such as moraines, ice-push ridges and outwash fans and indicate that, at maxima, ice coalesced across the North Sea. This happened for the first time in MIS 12 (Toucanne et al. 2009), during the Anglian glaciation, when British and Scandinavian ice sheets reached a maximum extent and covered all but the lower part of the Southern Bight. In comparison, Saalian ice cover, 250,000 years later, was less extensive over England but more extensive in the Netherlands. Over the southern North Sea region however, the Anglian and Saalian glaciations were comparable in coverage (Figs. 7.5 & 7.8).

During the Last Glacial (Weichselian), glaciation along the northeast English coast reached relatively far south. Over the central North Sea, however, Scandinavian ice reached only as far as the Dogger Bank, and areas such as the German Bight remained ice-free (Houmark-Nielsen & Kjær 2003). Even though the Scandinavian and British ice sheets may have converged over the North Sea, large tracts of its central and southern sections in the last 140,000 years have remained unglaciated (Fig. 7.5; Ehlers & Gibbard 2004; Clark et al. 2012). The Dover Strait region has never been glaciated.

During periods of maximum glaciation, large proglacial lakes are envisaged in the southern North Sea (Fig. 7.8), blocked by the ice-front to the north and upland topography to the south (Gibbard 1995; Toucanne et al. 2009; Hijma et al. 2012; Murton & Murton 2012; Cohen et al. 2014). The lakes are inferred/reconstructed both in the older Anglian and Saalian glaciations (parts of MIS 12 and MIS 6 respectively) and the younger Weichselian glaciation (Last Glacial, parts of MIS 2). Partial sedimentary evidence for these lakes exists. For the Anglian, water-laid ice marginal deposits are part of Norfolk's Anglian sequences (Lunkka 1994) and the Elsterian of the northern Netherlands contains subaqueous outwash fans of subglacially-entrained micaeous sands (Peelo Formation, Laban & van der Meer 2011; Lee et al. 2012). That this was part of one larger lake held up by a single ice-front (Fig. 7.8a) is inference, supported by the maximum ice-limit reconstructions for this period (Fig. 7.5).

For the Saalian (Fig. 7.8b), water-laid proglacial and ice-marginal deposits are known from the Cleaver Bank region (Laban 1995; Moreau et al. 2012) and the maximum-limit ice-marginal Rhine at initial stages grades to a high base level, inferred to be from the same lake (Busschers et al. 2008; Peeters et al. 2015a). Note that in these ice-limit zones, push-moraines and subglacial features were also created and the interrelations between these landforms, water-laid proglacial features, and water-laid deposits from deglaciation phases have long been difficult to separate in regional mappings (e.g. Laban 1995). Seismic interpretation appears to be changing this recently (e.g. Moreau et al. 2012). From the central North Sea, indications for water-laid proglacial and ice marginal deposition are reported in Laban (1995) for the central North Sea in the Dogger Bank region. Ephemeral formation and drainage of the Weichselian lakes to the Norwegian Sea is evident from δ¹⁸O signals in deep marine sediments (Lekens et al. 2006). We can infer that between sudden drainage events towards the north-west, these central North Sea lakes overspilled towards the south, and that their southern shores must be sought north of the Brown Bank follows from mapping of the Southern Bight region (Hijma et al. 2012).

Figure 7.7b shows an interglacial highstand situation with a half-eroded land bridge for the time period between MIS 12 and MIS 6. It highlights that a terrestrial connection continued to exist between East Anglia and the Netherlands, even after initial erosive cutting in the Dover Strait had begun (Funnell 1995; Hijma et al. 2012). This terrestrial connection was narrow and the saddle could potentially be overtopped by shallow seas during true interglacial highstands (i.e. as high as at present, if not a few meters higher), since for most of the MIS 12 to MIS 6 period, sea level was below such levels (White & Schreve 2000).

The position of the southern lake limits and the elevations of lake levels in Fig. 7.8 are tentative estimates. In part, the positions are supported by, for example, the distribution of estuarine sediments from the interglacial highstands that Fig. 7.7b summarizes as a single scenario (with further evidence from biostratigraphy: Meijer & Preece 1995; Roe & Preece 2011). The (initial) lake level elevation for the Anglian is supported by a major channel fill deposit at Wissant (Roep et al. 1975), interpreted to be a spillway fragment — but estimates of tectonic uplift and GIA at the time of and since the Anglian are also factored in. Similarly, the initial level at the time of the Saalian maximum is estimated, factoring in GIA and tectonic subsidence (Busschers et al. 2008).

That the ice sheets produced vast amounts of melt water at their maximum stage and that these gathered in ice-marginal rivers and were routed towards the North Sea is evident from onshore glacio-fluvial features, to which northward-flowing rivers from England and mainland Europe added further discharge. That major flows of water were routed south through the Dover Strait at lowstands, is evident from seafloor geomorphology of the English Channel (Gupta et al. 2007). The greater importance of glaciations during MIS 12 and MIS 6 in eroding the Dover Strait land bridge, compared to that of other glacials, is supported by sediment delivery rates as measured off the English Channel floor in records collected in deep marine settings (Toucanne et al. 2009).

The waters collected in the lakes shown in Fig. 7.8 were routed southward as lake overspill. This is used to explain the formation of the Dover Strait as the carving of a proglacial-lake spillway valley (Roep et al. 1975; Smith 1985). In turn, this would explain the signals of biogeographical insularity seen in paleoenvironmental records from Britain over a series of interglacials and give an indication of the first glaciation that was extensive enough to induce this change (Gibbard 1995).

The degree of importance attributed to proglacial spillage in opening the Dover Strait is a matter of interpretation. The explanations entertained here (based on Gibbard 2007; Gupta et al. 2007; Busschers et al. 2008; Hijma et al. 2012; Gibbard & Cohen 2015 — see Murton & Murton 2012 for an independent review) attribute maximum glaciation lake spillage a relatively large role in a short period (the Anglian and Saalian glacial maximum together span perhaps 10,000 to 20,000 years). Erosion by normal rivers and tributaries, estuarine processes and coastal cliff retreat are each attributed a smaller role despite being in operation over much longer time spans in the last 450,000 years. This is because both vertical erosion and the sudden carving of a new north-to-south valley have to be explained. Spillage would be the process to create a new path (as a headward cutting tributary could also do, but would take much more time), while river and estuarine erosion would be processes that deepen existing paths. Cutting of the spillway may have been catastrophic (Smith 1985; Gupta et al. 2007; i.e. an event within a glacial) or may have been gradual (i.e. going on for several thousand years during a glacial). At the time that the land bridge (sill) was >100 km wide (Figs. 7.7a & 7.8a), gradual retrogradation is more probable (i.e. in the Anglian), and by the time that the sill had shortened and lowered (Figs. 7.7b & 7.8b), a final catastrophic event is probable (i.e. in the Saalian).

In the aftermath of the Saalian deglaciation (the situation following Fig. 7.8b), the proglacial valley in the Southern Bight remained in use by the River Meuse. Discharge from the Rhine system had contributed to erosion of the Dover Strait land bridge at the times when proglacial lakes existed, but stopped doing so during the Saalian deglaciation (Busschers et al. 2008; Peeters et al. 2015a). As a lowstand river, however, it would only start to use the Axial Channel escape route out of the North Sea basin later in the Last Glacial (Fig. 7.9).

Details in this theory explaining the landscape evolution of the southern North Sea in the last 1 Ma have bearings on the taphonomy of the North Sea floor. They determine how major downcutting and/or reworking episodes operated in a given period. Hijma et al. (2012), for example, explore scenarios of valley downcutting at the end of the Saalian and during the Weichselian, that included trade-offs between proglacial spillway and periglacial river modes and intensities of erosion. Similarly, the depiction of the coastline configurations for the highstands between the Anglian and the Saalian (Fig. 7.7) would require scenario exploration of such trade-offs for these two glaciations (and possible further events in MIS 10 and MIS 8).

Late Pleistocene up to the Last Glacial Maximum

For the extent of the North Sea in the Last Interglacial and the situation of the North Sea floor in the Last Glacial (e.g. Fig. 7.9), abundant evidence is available that is independent of scenarios and interpretations regarding the erosional history of the Dover Strait in the Middle Pleistocene.

Figure 7.9a shows the situation of the southern North Sea during the Last Interglacial (MIS 5; Fig. 7.5). As a result of the cumulative effects of tectonic setting and progressive glaciations (see previous section, pages 154–159, and Figs. 7.5–7.8), the Last Interglacial was the first interglacial when Britain was an island (Bridgland & D'Olier 1995; Funnell 1995; Gibbard 1995; Meijer & Preece 1995). It was also the first interglacial when the North Sea had an extent and coastal configuration similar to today's (Streif 2004). At full highstand, the North Sea was a substantial barrier to biota, as expressed in terrestrial biostratigraphical differences (including hominin absence) between Britain and the Continent in the Last Interglacial (Funnell 1995). The Last Interglacial proper (regionally defined as the Eemian/Ipswichian; Fig. 7.4) was relatively short (just ca. 12 kyr, of which ca. 6 kyr was true highstand; Zagwijn 1983; Sier et al. 2015; Peeters et al. 2016). It was followed by slow sea-level fall with the Rhine Delta building out into the North Sea (Fig. 7.9a). In the later parts of MIS 5, this helped to create a lowland land bridge in the shallow southern North Sea allowing hominins to (re)occupy England from northwest Europe. A Neanderthal skull fragment found offshore of the southwest Netherlands, on grounds of morphological resemblance to such skulls in northwest France (Hublin et al. 2009) and the setting and inferred taphonomy of the offshore find location (Hijma et al. 2012), would be from this time.

As global climate cooled further into the Last Glacial and global sea level fell from ca. –25 m (MIS 5a, 85 ka) to ca. –90 m (MIS 4, 70 ka), an increasing expanse of the southern and central North Sea floor became exposed (Fig. 7.9b). Once again, river valleys traversed the exposed shelf and once again, the north of the basin became glaciated. It is not fully certain if the British and Scandinavian ice sheets coalesced in the northern North Sea, but it is suspected that they did for a brief period (Carr 2004; Carr et al. 2006; Lekens et al. 2006; Sejrup et al. 2009; Clark et al. 2012). The northern half of Denmark and the area of the Dogger Bank are home to large complexes of ice-pushed moraines that mark the limits of the Scandinavian ice sheet in the Last Glacial. British ice reached down along Lincolnshire to the north Norfolk coast, leaving tills and subglacially-carved lows such as the Outer Silver Pit (Fig. 7.1).

Parts of the Danish and German sectors may have hosted proglacial lakes for shorter intervals, with similar formative mechanisms to those of the Anglian and Saalian in the southern North Sea (see ‘Middle Pleistocene’ pages 154–159; Fig. 7.8). In this region, the Elbe ice-marginal river contributed water, as did outwash channels sourced from Scandinavia. At times of coalescent ice sheets and proglacial ponding, overflow of lake water must have been routed southward towards the river systems in the area of the present-day Southern Bight, Dover Strait and English Channel (Clark et al. 2012). Figure 7.9b shows possible spillage pathways joining fluvial drainage in the Southern Bight and the gorge of the Dover Strait. For the preservation of Early Glacial records in the southern North Sea, the lake spillage from the north may be relevant. The mapping of the pathways implies that large parts of the original regressive Rhine Delta Plain (Brown Bank Formation) have been eroded by waters from the north (compare Figs. 7.9a and 7.9b) and not just as a consequence of normal erosion by the Rhine. Recognizing northerly lakes and southerly spillage pathways could be important for taphonomic assessment of the contents of erosive lags (e.g. Glimmerveen et al. 2006) in fluvial deposits such as appear on the eastern side of Brown Bank (also see ‘Evidence of submerged landscapes on the shelf’ pages 166–169.

From the final stages of the Last Glacial (the last 18 kyr, if not the last 27 kyr), the Elbe ice-marginal system shows up as a valley system on the North Sea floor (Figge 1980; Houmark-Nielsen & Kjær 2003; Alappat et al. 2010), directed to the Norwegian Channel and graded to a base level of –70 m or deeper (Fig. 7.1). On the grounds of position, depth and grading to relatively low base levels matching contemporaneous sea level, this indicates that the ice stream from the Oslo fjord had withdrawn and no longer blocked the Elbe's northward drainage. Any Last Glacial major proglacial lake stage in the central North Sea must therefore date from before the functioning of that bathymetrically expressed last valley.

Late Glacial and Holocene

Postglacial inundation of the North Sea in the last 20,000 years, at a shelf scale, is typically visualized using output of GIA models (e.g. Lambeck 1995; Shennan et al. 2000; Vink et al. 2007; Sturt et al. 2013; Fig. 7.10). Besides numerical descriptions of the Earth's geophysical response to changing loads, GIA models use present-day bathymetry/topography, reconstructions of ice-sheet advance/retreat and suites of sea-level index points (SLIPs) as input data (see Milne 2015). The models then iteratively solve the values of properties describing the response of the Earth's crust and mantle in response to shifted loads of ice and water over the planet.

For ice-mass centers such as Scandinavia and Scotland the modeled GIA comprises subsidence during ice build-up and rebound during deglaciation, tailing out in postglacial times. For ice-sheet peripheral regions such as the central and southern North Sea (in this context dubbed ‘Near Field Regions’), it means upwarping during ice build-up and accelerated subsidence during deglaciation (e.g. Lambeck 1995; Shennan et al. 2000; Vink et al. 2007). Combined with the inland position of the southern North Sea on the continental shelf, this means that in each interglacial, this area is transgressed relatively late, but when it does, rates of sea-level rise are relatively high (‘drowns last, sinks fast’). This allows sufficient time for a boreal-to-temperate vegetation cover to develop and substantial postglacial soil formation to take place on the fresh surfaces of depositional landscapes inherited from the preceding glacial, before such surfaces are transgressed and drowned. This is a geographical property of the North Sea basin that affects the taphonomy of terrestrial surfaces buried and preserved by transgressive units. It is most evident for Late Glacial and Holocene drowned landscapes, for which radiocarbon dating provides an accurate independent age control on the timing of interglacial warming and marine transgression (e.g. Törnqvist et al. 2015), but applies equally to, for example, the basal surfaces of coastal plain units of the Cromerian Complex stage (e.g. page 156) and the Eemian stage (e.g. page 160).

GIA modeling results tend also to be expressed as a time-stepped series of maps that show the intersection of the sea surface with the warped topography/bathymetry (e.g. Fig. 7.10). In recent decades, GIA models have been improved and several generations of GIA-modeled North Sea drowning histories exist (e.g. Sturt et al. 2013 is an upgrade of Lambeck 1995). Both regionally and globally resolved versions of GIA models exist. Peltier (2004), for example, gives a global solution, whereas Lambeck (1995), Vink et al. (2007), Brooks et al. (2011) and Sturt et al. (2013) give regional solutions (see also Peltier et al. 2002; Steffen & Wu 2011).

Of the regional GIA-model solutions entertained for the North Sea, some optimize for the SLIP dataset around Britain (e.g. Lambeck 1995; Shennan et al. 2000; Peltier et al. 2002; Ward et al. 2006; Sturt et al. 2013), while others see the region as the periphery of Scandinavia and optimize using different SLIP datasets (e.g. Lambeck et al. 2006; Vink et al. 2007). Overlap in the SLIP datasets used also occurs, especially for offshore data points from the central North Sea. Furthermore, unlike the uplifted coastal areas of Scotland and south Scandinavia where the SLIP datasets cover both the Late Glacial (when rates of GIA were relatively fast) and Holocene (when GIA began to tail out), in peripheral areas to the south, SLIP datasets are biased to the Middle and Late Holocene. Efforts to synchronize the data sets in use by the various sea-level research, GIA modeling and offshore archaeology groups of the different countries around the North Sea are underway and include database protocol activities (Hijma et al. 2015).

GIA-modeled paleocoastline maps for the North Sea are sensitive to the choice of model setup and completeness of SLIP datasets that are used. In particular, for the Early Holocene (11–8 ka), there is considerable variation in the timing of drowning of the parts of the central North Sea now 20 m to 50 m deep, between different GIA-model studies — especially between areas that are spatially peripheral to one study but central in another. Dates that are collected for local geological and archaeological reasons should be used to verify the GIA model-predicted shorelines and, when related to timing of transgression, should feed into SLIP databases to improve future GIA-paleogeographical modeling.

Despite differences in approach, all GIA-modeled map series of recent decades show an initial period of relative stability in shoreline position around the northern and central North Sea between ca. 21 ka to 14 ka, when isostatic warping kept pace with rising glacio-eustatic sea level. The southern North Sea at this stage remained dry land and hosted river valleys that continued towards the English Channel. From ca. 14 ka onwards, marine transgression was rapid throughout, flooding the exposed land from the north and from the south via the Dover Strait. The most recent GIA modeling study calculates that the terrestrial connection between Britain and Europe began to be overtopped at ca. 9.5 ka and the final large island in the central North Sea (the present Dogger Bank) was submerged by ca. 8 ka to 7.5 ka (Sturt et al. 2013; Fig. 7.10).

Independent of the SLIP datasets used in the GIA modeling, marine sedimentary indicators show the establishment of mixed southerly and northerly sources of water in the Skagerrak by ca. 8.5 ka (Streif 2004; Gyllencreutz 2005). This indicates when water depths in a corridor between the southern and central North Sea had increased enough to allow establishment of current systems similar to today (see also van der Molen 2002) and thus when marine erosion and reworking processes that were the final taphonomic control on preserved drowned landscapes and archaeological sites began.

From 7.5 ka, Fig. 7.10 shows the shape of the North Sea as similar to that today. In part, this is an artefact of the bathymetric/topographic data used as input to generate the maps, which show recent features that simply did not exist during the modeled time steps. For instance, these data include the morphology of beach-barrier systems that formed in the last 7.5 kyr. The coastal zones of the southern North Sea, in particular the sandy shores of Flanders, Zeeland, Holland, the Waddenzee and the German Bight show backstepping of the coastal system until 6.5 ka (southwest Netherlands: Beets & van der Spek 2000; Hijma et al. 2010; Hijma & Cohen 2011) or ca. 5 ka (Waddenzee, German Bight: Oost 1995; van der Spek 1996; Streif 2004; Vos 2015) and maturing barrier systems thereafter. Major parts of the presently barrier-protected Waddenzee and Dutch coastal plain were, at 7.5 ka, exposed to the open North Sea and the true coastline was positioned inland of that depicted in Fig. 7.10.

These effects are also relevant to the western side of the southern North Sea. For instance, although the model does account for Late Holocene infilling of the low-lying East Anglian fens (by using borehole-derived isopach maps instead of modern topography: Sturt et al. 2013), it does not incorporate erosion and retreat of the unconsolidated cliffs along the outer coast of East Anglia (e.g. Dong & Guzzetti 2005). For the coastal zones of the North Sea, geological mapping and paleogeographical reconstruction documenting their evolution is well developed, but incorporation of such mapping in GIA-modeled shorelines to cover not only the offshore, but the full North Sea, is not (Cohen et al. 2014). This most affects taphonomic assessments for the period between 8 ka and 3 ka, especially in nearshore, tidal inlet and estuarine areas. For taphonomy farther offshore, the GIA-modeled visuals for the North Sea are of more direct use.

Because of the gentle gradient of the shallow shelf, a stepwise acceleration between 8.45 ka and 8.25 ka (as initiated by the meltwater pulse triggering the 8.2 ka Event) may have been of particular importance to the drowning of the central North Sea at the turn of the Middle to Late Mesolithic. The onset of this sea-level rise acceleration is radiocarbon dated in transgressed terrestrial sediments in the southern North Sea off and below Rotterdam (Hijma & Cohen 2010). The structure of this sea-level jump, which occurred on top of a background rate of about a meter of rise per century, reveals that at two moments within these two centuries, sea level jumped by a meter (because of events on the other side of the Atlantic Ocean, where Canadian lakes Agassiz and Ojibway drained as Hudson Bay and the Tyrrell Sea became ice-free; Teller et al. 2002). Along the shores of the contemporaneous North Sea (e.g. Vos et al. 2015: Maasvlakte area, off Rotterdam harbor), a meter of sea-level rise would have occurred within the lifespan of a human generation (say 20–25 years), and another such meter-scale jump about 150 years later. In discussions of coastal Mesolithic archaeology, transgression as a push factor to inland migration and contact between coastal and terrestrial Mesolithic societies is often mentioned (e.g. Smith et al. 2011). Superimposed events such as the 8.45 ka to 8.25 ka twinned sea-level jump could have catalyzed this worldwide (e.g. Turney & Brown 2007) and in the North Sea, the Storegga tsunami occurring ca. 8.1 ka (Rydgren & Bondevik 2015) could have had a similar impact (e.g. Dawson et al. 1990; Weninger et al. 2008). In the context of this book and chapter, it is the impact on preservation and taphonomy of the superimposed transgression events that should be noted. For the Storegga tsunami, this could be considered one of destruction, but for the sea-level ‘jumping’ in centuries before it appears to have improved preservation of surfaces (Hijma & Cohen 2011) and actual archaeological sites (Vos et al. 2015), at least in the southern North Sea nearshore paleovalley settings that were transgressed by the event.

Given the role SLIPs play in GIA-modeling in determining regional modeling outcomes and predicted paleogeographies, their collection and availability is particularly important. The sides of drowned paleovalleys in particular provide suitable circumstances for collecting series of SLIPs (Vis et al. 2015). Furthermore, it is useful to group offshore SLIPs for the North Sea by parent paleovalley system/drainage network, because valleys were transgressive pathways. Although an estimated 400 suitable SLIPs are presently available from the British, Belgian, Dutch, German and Danish North Sea sectors and coastal zones, these mainly cover the Middle and Late Holocene depth intervals (final 15–20 m of sea-level rise; e.g. Shennan et al. 2006). Data points for 14 ka to 7.5 ka, obtained from offshore areas where the North Sea is 20 m to 60 m deep, are much more scarce (uplifting coastal areas of Scotland and Norway excluded). Shennan et al. (2000) and Hazell (2008) report 15 to 20 such data points for the British sector, and some 20 more from the Dutch and German sectors collected in the 1970s. With recent archaeological dating campaigns in the various wind farm localities, the number of offshore collected SLIPs is now increasing on the British side. Vink et al. (2007) list another 60 offshore data points from the Dutch and German sectors combined (Fig. 7.11) and Alappat et al. (2010) add five more data points from the offshore Elbe paleovalley.

Besides collecting SLIPs from the offshore, typically from below 25 m depth, collecting them from the nearshore region is critical to cover the interval between 15 m and 25 m deep. All along the southern North Sea — the English coast, the Dutch-Belgian shores, the Frisian Islands (Waddenzee) — this is the depth range where 9 ka to 8 ka sea levels are projected, but where data points are particularly scarce. The reason is that deposits at this target depth in these coastal/nearshore environments are difficult to survey and core either from the sea or the land. From the sea, the difficulty arises firstly because at water depths of just a few meters seismic techniques investigating the target depth suffer from multiples of the water/sediment contact, and secondly because to reach the target transgressed surfaces, meters of recent shoreface and coastal sediment have to first be cored. From the land, depths of 15 m to 25 m are cored infrequently. Regionally along the North Sea this makes the nearshore area a ‘white zone’ for direct data on submerged landscapes, and has notably affected SLIP datasets too. Exceptions to this are areas with harbors and concentrations of economic and building activity, such as the Rotterdam harbor complex that overlies the Early Holocene valley of the Rhine and Meuse. Hijma and Cohen (2010) and Vos et al. (2015) added ten SLIPs from this critical 15-m- to 25-m-deep interval.

Relative sea-level rise, due to residual GIA and background tectonic subsidence combined, has raised the seawater surface of the North Sea further above the seabed by ca. 5 m to 8 m in the last 6 kyr, compared to a rise of 15 m to 35 m between 9 ka and 6 ka (lower values for the south than for the north, see Fig. 7.11). In the last 6000 years, sedimentary processes in the shallow sea (creating offshore sand-wave fields, shore-connected ridges, foreshore and beach bar systems, tidal inlets) controlled the preservation and taphonomy of drowned and buried landscapes of the nearshore zone (see ‘Physical geography and modern environment’, pages 150–152).

Outlook on data, mapping and reconstruction quality

Before discussing the submerged landscape inventory and taphonomic ramifications of the above, it should be noted that the overview presented here represents only the broadest picture, whereas dedicated original studies at subregional scale provide much more detail (e.g. Hijma et al. 2012; Tizzard et al. 2014). Furthermore, paleogeographic reconstruction inevitably involves a degree of uncertainty and interpolation, particularly where there are gaps in the geological record caused by erosion as well as uncertainties in the dating and spatio-temporal correlation of the parts of the record which are preserved — and the examples from the North Sea show this (Cohen et al. 2014). Inevitably, the gaps and uncertainties are greatest for the earlier parts of the record, generally reducing as the data get younger. Thus, for the Middle Pleistocene, paleogeographic and paleolandscape changes are only reconstructed on a basin scale, usually as rather generic highstand/lowstand scenarios with considerable uncertainties in the timing of events and exact position of features such as rivers and coastlines. For the Late Pleistocene and the post-Last Glacial Maximum (LGM), the data allows time-stepped reconstructions at millennial intervals to be produced, ideally by combining information from GIA-modeling results and geological reconstructions, including those based on seismic data (e.g. Gaffney et al. 2007; 2009; van Heteren et al. 2014).

In the best-case scenario, detailed investigations combining multidisciplinary data and secure chronology can provide a more refined reconstruction which includes the identification of specific landform assemblages and also accounts for changes in erosion and deposition patterns (Bicket 2013; Cohen et al. 2014). For the Dutch sector, Erkens et al. (2014) assessed whether the current wealth of data (Figs. 7.2 & 7.3) would allow reconstructions of the submerged terrestrial geology offshore comparable to those for the onshore. It was concluded that the desired level of detail of paleogeographic reconstructions suitable for archaeological prediction could not be achieved without re-interpreting most of the existing data and carrying out additional fieldwork at considerable cost. In this light, it is good that ca. £60 million (approx. $73 million) is being spent on surveying and understanding the British part of Dogger Bank alone, as part of wind farm planning and development (Forewind consortium; www.forewind.co.uk).

Figure 7.2 (A) Coverage of paleolandscape projects (gray polygons) in the UK sector of the North Sea (from Bicket 2013: fig. 8). Map includes the ‘3D seismic central North Sea Mega Merge’ area of Gaffney et al. (2007); (B) Coverage of 3D seismics in the Dutch sector of the North Sea (www.nlog.nl) (from Erkens et al. 2014). Note that many of the seismic cubes, especially in the south, are not suitable for the identification of shallow preserved landscapes, due to the shallow water depths. Colors indicate the year of collection. In the first 5 to 7 years after collection, the collected data are often confidential (red blocks date to 2008).

Figure 7.3 (A) Example of density of borehole data, and; (B) 2D seismic line data in the Dutch sector of the North Sea as available from the databases of the Geological Survey of the Netherlands. Erkens et al. (2014).

Figure 7.4 Correlation chart for the North Sea basin for the last 1 million years, showing chronostratigraphic stages and terms. Modified from Gibbard and Cohen (2008), with additional archaeological information from Pettitt and White (2012).

Evidence of Submerged Landscapes on the Shelf

The North Sea holds a range of archaeological, sedimentary and geomorphological evidence for submerged prehistoric landscapes, i.e. drowned preserved former land surfaces, conserved and not substantially eroded or reworked by younger landscape and seabed-forming processes. This evidence has been found in different parts of the sea basin at a range of depths (both water and burial depth) reflecting patterns in preservation and taphonomy. Detection and reconstruction of buried elements of the submerged paleolandscape has, in recent years, been greatly aided by the provision of vast seismic data sets, often originally collected for hydrocarbon prospection (e.g. Gaffney et al. 2007; 2009; van Heteren et al. 2014). For a full review of the inventory of features composing the former landscapes in the North Sea, see Cohen et al. (2014). Reviews of the archaeological evidence can be found in Peeters and Momber (2014) and Roebroeks (2014).

Geomorphological evidence includes a range of glacial and periglacial landforms. The geological literature pays considerable attention to the North Sea's many subglacial tunnel valleys. These have been found across the sea basin, principally to the north of the Southern Bight, from all glaciations covering the central North Sea (Huuse & Lykke-Andersen 2000; Praeg 2003; Stewart et al. 2013). Upon deglaciation, the tunnel valleys developed into lakes, coastal embayments or local deeps in a shallow shelf sea and were accordingly filled with whatever sediments they trapped. Because of their transformed state, the tunnel valleys are of archaeological landscape relevance.

An example for which the postglacial archaeological potential has been explored is the Outer Silver Pit (Fig. 7.2; Fitch et al. 2005; Fitch 2011). These studies highlight that the shores of larger freshwater bodies would be attractive base-camp locations for terrestrial hunter-gatherer communities. Furthermore, tidal conditions were in play early on in the transgressive episode (Fig. 7.10; 10.5–10 ka BP), which established a depositional environment that could cap and protect former lake shores from erosion in later stages. Tunnel valleys are of further relevance as seismo-stratigraphic marker contacts that can be regionally traced through 3D seismic volumes and allow the distinction of Pleistocene sequences and associated relative ages. Other currently identified glacial/periglacial landforms include iceberg scours (particularly in the northern North Sea), till sheets with mega-scale glacial lineations (north of maximum ice limits), and moraines, ice-pushed ridge complexes and ice-marginal river valleys in the maximum ice-limit regions (Busschers et al. 2008; Moreau et al. 2012).

Preserved river valley floors (floodplains and channel fills) from periglacial and interglacial landscapes form a second major component of North Sea paleolandscapes. Some of these are particularly well-traced using 3D seismic data in the central North Sea in the vicinity of the Dogger Bank and Outer Silver Pit (Fig. 7.12; See Gaffney et al. 2007; 2009; van Heteren et al. 2014). Most of the mapped channels from this area are inferred to be of Late Glacial and Early Holocene age (where they connect to the Outer Silver Pit, which is itself of Devensian age), while older examples can also be seen in the data at greater depths (Fitch et al. 2005). Paleochannel fills and their channel belts have also been identified off the Netherlands (Hijma et al. 2012), northern Germany (Konradi 2000; Streif 2004), Belgium (Mathys 2009) and East Anglia (Bridgland et al. 1993; Bridgland & D'Olier 1995; Dix & Sturt 2011; Tizzard et al. 2014). Other identified interglacial features include lakes, marshes, floodplains, estuaries (with tidal channels), inland eolian dunes, and various nearshore sedimentary bodies (Hijma et al. 2010; Cohen et al. 2014; van Heteren et al. 2014). In general, basal parts of these features are better preserved because of erosive truncations of their tops since transgression (tidal scour, wave ravinement — see ‘Physical geography and modern environment’, pages 150–152).

Over considerable areas offshore, because the Early Holocene transgression was rapid (Fig. 7.10) and rates of sea-level rise high (Fig. 7.11), the odds for preservation of patches of terrestrial landscape offshore may be better than in the coastal zone of the Middle Holocene (due to tidal scouring of inlet channels), especially where depositional terrestrial environments were transgressed. In general, Early Holocene valley surfaces would be expected to preserve better (intact soils) than the surfaces of their interfluve areas (ravinement decapitated). Moreover, even if the Early Holocene valley surface was transgressively eroded, the truncated valley deposit may have buried a Late Glacial precursor surface, so there would still be a chance that a paleolandscape surface is preserved and in situ archaeology encountered.

Some of the aforementioned paleolandscape elements have been identified from sedimentary (i.e. lithological, biostratigraphical) rather than geomorphological evidence, generally obtained from core or borehole samples. While the geomorphological evidence, derived principally from seismic and to a lesser extent bathymetric data, is well suited to regional-scale landscape mapping, the sedimentary record is essential in providing paleoenvironmental (e.g. freshwater vs. marine sediment) and dating evidence that allows a more accurate paleolandscape reconstruction, and is crucial when trying to reconstruct change over time (van Heteren et al. 2014).

The most immediately obvious paleoenvironmental resource comprises submerged peats formed upon flooding of the terrestrial landscape (Arends 1833; Reid 1913). In situ basal peats are compacted peats, buried by tidal brackish deposits marking transgression (Jelgersma 1979; Vis et al. 2015). Throughout the central and southern North Sea, these date to the Late Glacial and Holocene depending on depth and geographical position (Fig. 7.11). They have been found at various localities including the Dogger Bank (Ward et al. 2006; Hazell 2008; Vink et al. 2007; Alappat et al. 2010; Hijma & Cohen 2010). Older examples have been reported, for instance, as part of the top of the Yarmouth Roads Formation (Zagwijn 1979; Funnell 1996). These older peats represent the remnants of the coastal plain transgressions from the Cromerian Complex (see Fig. 7.4) and have been correlated, based on stratigraphical position and contained palynology, with the onshore formations that contain in situ Paleolithic artifacts at Pakefield and Happisburgh (Parfitt et al. 2005; Wessex Archaeology 2009; Bicket 2013; Tizzard et al. 2014).

Overall, a range of other identified sedimentary deposits representing different environments and periods ranging from the Middle Pleistocene to the Holocene includes tidal flat, periglacial eolian, lagoonal/lacustrine, floodplain, channel infill, river-valley inland dune, marine, glacio-lacustrine and glacio-fluvial deposits (Cameron et al. 1992; Gatliff et al. 1994; Balson et al. 2002; Hijma et al. 2012; Cohen et al. 2014; Tizzard et al. 2014). Besides peat beds, it is negative relief landforms which host lakes and collect lacustrine fills that provide excellent paleoenvironmental records, which is an extra reason to map (Fig. 7.11) and target them for coring (e.g. Tizzard et al. 2014).

An aspect which remains under-researched is vegetation reconstruction. So far, pollen studies from these submerged environments are few and far between; hence, vegetation reconstructions are reliant on data from land which may be over 100 km away (Peeters & Momber 2014). As an illustration, the Late Glacial and Early Holocene vegetation developments in Doggerland's northerly interfluve landscapes, between the major Elbe (German Bight) and Rhine-Thames (Southern Bight) paleovalleys, could be expected to deviate from those in the far west of onshore northern England, the far south of onshore Netherlands and Germany, and the far east of Denmark and Sweden. For example, the timing of the invasion of a key tree species such as Corylus (hazel) into Doggerland is not yet directly established. Corylus is important because its nuts were a major Mesolithic food resource and charred remnants are often indicative of human activity or settlement. The appearance of Corylus in pollen records also marks the Preboreal-Boreal transition in the regional pollen zone scheme. Direct correlation to onshore areas such as the eastern Netherlands (van Geel et al. 1980) would place this transition at ca. 10.25 ka (ca. 9125 ±90 ¹⁴C BP based on radiocarbon dates from the eastern Netherlands), but it might be a few centuries later in the Dogger Bank region since it lies some 300 km further north.

The aforementioned deposits and landforms are the evidence of the former landscape, and provide fossil material indicative of the various paleoenvironments of the time (interglacial vs. glacial flora and fauna, terrestrial vs. aquatic species, freshwater vs. marine biota). In principle, this also applies to archaeology (e.g. harpoons encountered from wetland and coastal environments) although hunter-gatherers may also have moved resources between environments, especially in coastal and wetland environments (Sturt 2006; Waddington et al. 2007; Amkreutz 2013; Bell et al. 2013; Moree & Sier 2014; 2015; Peeters et al. 2014; 2015b). Concentrations of faunal remains appear to be in the same areas as the archaeological remains, coming principally from the Brown Bank and Eurogeul navigation channel (five nautical miles west of Rotterdam harbor), with additional material from the Zeeland Ridges, the Belgian coast and off East Anglia (van Kolfschoten & van Essen 2004; Mol et al. 2006; Hublin et al. 2009; Peeters et al. 2009; SeArch 2014; Tizzard et al. 2014).

The vast majority of finds from the North Sea are paleontological, comprising a range of Pleistocene and Holocene fauna such as mammoth, reindeer, horse and bison. To date, thousands of fossil bones have been landed, principally in the Netherlands, by trawlers and dredgers (Glimmerveen et al. 2004; 2006; Mol et al. 2006; van der Plicht & Palstra 2016). On paleontological grounds, and for the younger assemblages by radiocarbon dating, the trawled bone finds fall into four broad periods (Fig. 7.4): Early Pleistocene, early Middle Pleistocene (Cromerian), Late Pleistocene (Weichselian) and Holocene, with the majority dating to the latter two intervals (van Kolfschoten & van Essen 2004). It is worth noting however that the Late Glacial Period, as with the archaeological record (discussed below), is under-represented with only two radiocarbon-dated finds falling within the period between the LGM and the earliest Holocene (ca. 20–10 ka) versus more than 50 radiometric dates from the pre-LGM and Early Holocene periods (Peeters & Momber 2014). In recent papers, the limits of radiocarbon dating of bone material (van der Plicht & Palstra 2016) and of shell material (Busschers et al. 2010) from within Late Pleistocene strata of the North Sea are under discussion.

Some areas and formations, based on collections over the past decades, have provided far greater numbers of fossils than others. For instance, van Kolfschoten and van Essen (2004) specifically identify the Yarmouth Roads Formation (Early-Middle Pleistocene deltaic to non-marine), the Brown Bank region (Fig. 7.9a) and the Rhine-Meuse paleovalley channel sands and floodplains (Fig. 7.9b) as the source of fossil remains (see also Ward & Larcombe 2008; Hijma et al. 2012). Where these fossils are trawled from the actual seabed, marine reworking and winnowing processes are typically considered to explain part of the apparent concentration of fossils at the sea floor (whether in the last 6000 years by wave action, or in earlier stages of marine transgression). That said, it must be remembered that each of these areas encompasses reworking fluvial channel environments as well as floodplains, and that they formed over periods with oscillating climate and sea levels so that multiple phases of reworking are part of these units. Kuitems et al. (2015) provide an example from a sand extraction location off Rotterdam harbor as a local study, and Hijma et al. (2012) do so for the southern North Sea as a regional study. Local and smaller regional studies show that the context of finds from within subsurface units can only be revealed by detailed geological investigation. Doing so in more places, and also farther offshore, would be a step forward in improving assessments of taphonomy and paleogeography (e.g. Tizzard et al. 2014; Kuitems et al. 2015).

The evidence for hominin occupation of the North Sea presently comprises a variety of artifacts and bones trawled/dredged from the seabed. At present, the earliest known archaeological finds are Middle Paleolithic and include lithics from Area 240 (Fig. 7.3) off East Anglia (inferred date: Middle Saalian; ca. MIS 8–7: Tizzard et al. 2014) and the Zeeland Ridges off the Netherlands, the same area of the sea floor which has also provided a Neanderthal skull fragment (Hublin et al. 2009; Peeters & Momber 2014). The potential for earlier finds is suggested by trawled/dredged Early and Middle Pleistocene fauna and the presence of flint artifacts and footprints in Cromer Forest-bed sediments exposed on the lower beach and foreshore at Happisburgh, at the base of the eroding cliff (Ashton et al. 2014), of which extensions could also be preserved offshore, within reach of aggregate mining and wind-farm foundation activities (Wessex Archaeology 2009; Cohen et al. 2012; Bicket 2013; Ward et al. 2014).

Upper Paleolithic finds are rarer and comprise a barbed point from the Leman and Ower banks trawled up in 1931 (radiocarbon dated to 11,740 ±150 ¹⁴C BP: Housley 1991) and a possible worked flint from the Viking-Bergen Bank in the far north of the sea basin (Long et al. 1986; Coles 1998; Peeters & Momber 2014). The majority of finds to date are Mesolithic and include human remains, lithic, bone and antler artifacts. Relatively many human remains, as well as some of the bone and stone implements, come from the De Stekels area south-west of the Brown Bank.

The Maasvlakte-Europoort area off Rotterdam has produced over 500 bone and antler implements including barbed points (Louwe Kooijmans 1975; Verhart 2004; Peeters & Momber 2014). Also the Eurogeul navigation channel has also produced bone artifacts dating to the Mesolithic (Glimmerveen et al. 2004; Mol et al. 2006). Most recently, excavations in the second Maasvlakte harbor extension (Yangztehaven) have led to the discovery of an in situ Mesolithic assemblage at ca. 20 m depth, recovering thousands of bone fragments (including many charred bones), charcoal, lithics, fish and plant materials (including charred plant materials; Moree & Sier 2014; 2015 and contributions therein). This material was found by systematic archaeological sampling, guided by the results of detailed geological mapping (Vos et al. 2012; Weerts et al. 2012; Borst et al. 2014) starting from established paleogeographical understanding of the geology and archaeology of the region (Louwe-Kooijmans 1975; 1980; 2005; Hijma & Cohen 2010; Vos et al. 2011; Vos 2015), including the taphonomy. That taphonomy for the isolated inland dune upland within the Rhine–Meuse wetland involves post-depositional processes such as decay/selection, horizontal and vertical displacement and diffusion/blurring of the archaeological assemblages which collect in the wetland zones at the foot of local uplands (Amkreutz 2013:75), and the way further depositional developments bury, seal, compact and further seal and bury the artifact-bearing strata as transgression proceeds (Vos et al. 2015). From a sand-extraction area just offshore of the in situ Mesolithic assemblage, human skull remains were found (Borst et al. 2014), for which radiocarbon dating produced an age matching that of the Yangtzehaven base camp (Weerts et al. 2015).

Finally, post-Mesolithic finds have also been recovered, in the form of Neolithic axes from the Brown and Dogger banks (Peeters et al. 2009; Peeters & Momber 2014). The last shallow parts of the southern North Sea and the Dogger Bank area drowned by 7.5 ka to 7.0 ka (Fig. 7.10), just prior to the Mesolithic to Neolithic transition in the Netherlands and England. At the present state of research it is unclear whether these Neolithic axes could indicate in situ sites on former islands in the North Sea.

Taphonomy

Taphonomic variables

In ‘Quaternary background and paleogeographic framework’ and ‘Evidence of submerged landscapes on the shelf’, links were made between paleogeographical circumstances and developments and the taphonomic variables that are summarized in this section. Similar to paleogeography, the background controls on landscape preservation — and therefore archaeological preservation and site taphonomy — are tectonics, glaciation and sea-level change. The continuous landscape changes over multiple glacial-interglacial cycles described above mean that landscape-formative and record-preserving processes and conditions varied spatially through the period of interest.

Over the 500,000 to 1 million years covered by the archaeology of the region, this has created a very complex and fragmentary geological and archaeological record. A first step to organise it is to consider what deposits should be considered locally to contain archaeology (either in situ or concentrated reworked assemblages) and what not, from a climate perspective and past hominin climatic/environmental tolerances. For the Middle Pleistocene time frame, preserved deposits relating to interglacial (or even early/late stage glacials before/after maximum cooling) land surfaces are most relevant, given the archaeological information on climate tolerances of the hominins occupying the North Sea area and immediate surroundings at that time (e.g. Pettitt & White 2012). From the Late Pleistocene onwards, archaeology can be expected to be associated also with preserved surfaces within deposits of colder periods, with boreal forest, steppe and steppe-tundra environments alternating in the uplands, with the Rhine-Meuse and Thames river valleys with associated vegetation running through them (Fig. 7.9b). Throughout the Holocene, the North Sea area from a climate perspective was habitable.

A second step is to consider the controls on spatial patterns of preservation, in order to characterize regions based on the time depth of the archaeology that might be encountered in the first meters below the seabed (e.g. the first 10 m). Once again, these controls are tectonics, glaciation history and sea-level history primarily. These have been discussed above in relation to the Cromerian Complex of the Southern Bight (early Middle Paleolithic, pages 154–159). Crustal uplift/downwarping driven by long-term tectonics and glacio-isostasy is a taphonomic variable for younger deposits too, but becomes less influential with the increasingly direct impact of glaciation.

Sea level, glaciation, glacio-hydro-isostatic and climate history are strongly correlated in their cyclicity with ca. 100 kyr periodicity in the last 1 Myr. These correlations and interrelations make this second step of listing the taphonomic variables for the North Sea more complex than the first one. Also, for the North Sea, because of the vicinity to the Scandinavian ice-mass, tectonic basin longevity, and shelf size and gradient, this may work out differently compared to surrounding shelf areas and marine basins (i.e. English Channel, Baltic Sea). See Cohen and Lobo (2013) for a global perspective on the alternating morpho-sedimentary processes and human habitats on shelves.

With regard to tectonics, subsidence promotes sediment accumulation, which in turn affords preservation through burial. Since accumulation requires a supply of sediment as well as accommodation space for deposition, the location of depocenters represents a critical taphonomic variable. The depositional pattern also means that deposits are stacked vertically with younger ones overlying their older counterparts. Given the huge supply initially provided by the Eridanos River, the majority (ca. 80%) of the Quaternary succession — in southern depocenters over 500 m, towards the north over 1000 m thick — comprise shallow marine pro-deltaic and deltaic deposits of essentially this river system (Cameron et al. 1992; Funnell 1996). In contrast, on the basin shoulders where subsidence is minimal, preservation tends to be poor owing to reworking of unburied sediments. In special situations though, preservation is possible, and the minimal subsidence results in a complex mix of deposits of different ages located at roughly the same level.

This latter situation characterizes the Cromerian Complex sediments of East Anglia which are a composite of coastal highstand deposits of different ages interspersed with discontinuous transgressive and fluvial sediments respectively laid down during periods of sea-level rise and fall. Preservation of interglacial deposits here was enabled firstly by sediment supply at times with prolonged slow rates of sea-level rise which promoted coastal aggradation. Secondly, relatively consistent paleogeography promoted superposition of coastlines during successive interglacials (see Fig. 7.7a). Thirdly, restricted extents of ice sheets prevented direct glacial erosion and when glaciation did occur in MIS 12/the Anglian (Fig. 7.4), it buried the coastal plain deposits under a thick till sheet rather than eroding them, owing to waterlogged ice-front conditions (see Fig. 7.7b and Cohen et al. 2012).

It should be realized that rates of GIA (see ‘Late Glacial and Holocene’, pages 161–164) are an order of magnitude greater than long-term tectonic subsidence rates of the North Sea basin (Kooi et al. 1998) and that glaciations of the central North Sea have left positive topographic features that need multiple glacial-cycles to be topographically leveled out or sink away in sedimentation-filling subsidence-created accommodation space in the immediate surroundings. Besides controlling rates of subsidence and relative sea-level rise in times of postglacial transgression, during the part of the glacial interval that the North Sea floor was terrestrial, GIA in tandem with background tectonics affects the morphology of terraced valleys (Maddy et al. 2000). It is also part of the reason why considerable valley incision is seen over the tectonic depocenters (Fig. 7.5), as features in the 3D seismics from the central North Sea confirm (Fig. 7.12). The continued effects of tectonics and isostasy mean that one should not expect age-equivalent deposits to be situated at the same depths.

Figure 7.5 Location of tectonically subsiding areas (depocenters) in the North Sea region, and maximum limits for the Weichselian, Saalian and Anglian/Elsterian glaciations (see Fig. 7.4). Hijma et al. (2012).

Figure 7.6 Drainage basin and delta of the Miocene–Pliocene–Early Pleistocene Eridanos river system. Background map depicts former terrestrial lowland environments in Middle Pleistocene glacially-excavated areas that are nowadays occupied by seas. Reproduced from Cohen et al. (2014).

At times of upwarping towards the glacial maximum, GIA amplifies the depth of incision, whereas during deglaciation it amplifies the thickness of the eventual valley fill as the area subsides (Busschers et al. 2007; Hijma et al. 2012). Furthermore, the lower reaches of a tributary system tend to grade to the larger trunk river that they join, which has more stream power and erosive capacity. Submerged reaches of the Thames, Medway and Scheldt in the vicinity of the Axial Channel through the Dover Strait (Fig. 7.8b, Figs. 7.9a,b) thus have steeper gradients and vertically more pronounced terrace staircases than inland reaches. Submerged terraces have been identified offshore from bathymetric and seismic data (e.g. Bridgland et al. 1993; Bridgland & D'Olier 1995; Hijma et al. 2012; Fig. 7.9a), though care should be taken in attempting correlations between onshore and offshore terrace systems (Bates et al. 2007). These two glaciation-related particularities of the river systems of the North Sea — GIA effects and proglacial pathway inheritance — could have bearings on the taphonomy of preserved surfaces and their contained sites (e.g. Pettitt & White 2012).

A further taphonomic variable for river valleys, directly related to the tectonic setting and incisional vs. aggradational regimes, is the substrate composition of valley floor and valley side. These differ greatly between the seabed over depocenters off the northwestern Netherlands (Fig. 7.5) where the substrate is mainly sand (apart from the occasional patch of glacial gravels), and the seabed between Flanders and Sussex in the Southern Bight where gravelly sand (including abundant flint) occurs. This availability of gravel-sized lithologies in former terrestrial and coastal landscapes would affect tool-making hominins, particularly in the Paleolithic and could be an additional control on site distribution (Cohen et al. 2012; Bicket 2013).

Following the increased intensity of climate oscillations, starting 1 Ma and becoming more direct by ca. 450 ka, glaciations and sea-level oscillations increasingly overprinted the taphonomic contribution of tectonics. The direct impact of glaciation is generally regarded as negative, resulting in the erosion and reworking of the pre-glacial sedimentary record. This is most notable in the relatively rarity of in situ interglacial landscapes compared to glacial and periglacial deposits since they tend to be reworked or destroyed by the subsequent glaciation (Cohen et al. 2014). Large glaciation tends to be destructive. Exceptions can of course occur and certainly did in East Anglia (see pages 154–159), but only immediately south of the margins of the glaciation at maximum limit. This also means that the preservation of post-LGM (i.e. Late Pleistocene/Holocene) deposits is enhanced relative to earlier periods as they have not yet been affected by subsequent glaciation. On the other hand, erosional features created by glaciation can form sinks (e.g. tunnel valleys, kettle holes) in which later interglacial or postglacial deposits can accumulate. Water-laid sequences accumulated in such depressions could provide excellent paleoenvironmental information, but the presence of archaeological evidence in these sequences is unlikely unless lakeshore/bankside sediments are also preserved (see also ‘Evidence of submerged landscapes on the shelf’, pages 166–169).

Outside the ice margins, coversand/loess deposition, enhanced in rate and extent by periglacial conditions, can provide a protective blanket for archaeological sites and associated landscapes. This has been implicated in the potential differential preservation of Paleolithic sites in Britain relative to their continental European counterparts during MIS 6 to MIS 4 (Hijma et al. 2012). The effects are however unclear for the exposed North Sea landscape as it may have been the source, rather than a sink, of the eolian sediment present in the surrounding uplands (see also Antoine et al. 2009).

Sea-level change affects preservation by shifting the erosive effects of waves and tides across the continental shelf. That said, the actual degree of impact is also dependent on the rapidity of sea-level change (at the rates seen in the Late Glacial and Early Holocene, with accelerations due to meltwater pulsing), the nature of the coastline in question and the local sedimentary regime (Flemming 2004). For instance, back-barrier lagoons and the inner parts of estuaries tend to experience lower energy regimes compared to exposed outer coasts, while significant deposition driven by fluvial supply and/or longshore transport can promote preservation through burial. Growth of ‘basal’ peat induced by rising sea level can also provide a further means of protecting surfaces in the run up to true transgression (submergence of the peat surface), when the submerged peat is buried by mud and compacts. In particular, submerged deltaic deposits from transgressive stages in the valleys of larger rivers may contain and cover intact terrestrial surfaces with in situ archaeological remains. This is the case over large areas in the Rhine-Meuse paleovalley offshore, nearshore and onshore (Hijma & Cohen 2011; Hijma et al. 2012; Vos et al. 2015) and would also apply to the Elbe valley (Figge 1980; Vink et al. 2007; Alappat et al. 2010). Transgressive deposition could be especially efficient in burying valley floors in areas where tributary valleys joined (Vis et al. 2015). This appears to hold for some of the nearshore reaches of the smaller estuaries surrounding the Thames Estuary, such as the Medway, which has preserved submerged basal sequences from the Early Holocene along its fringes (Devoy 1979).

Of the taphonomic variables discussed above, some of them combine in peculiar ways. Postglacial marine transgressions, for example, use valley systems as the pathways to enter the North Sea. In the southern North Sea they have gradients controlled by the erosional history of the Channel through the Dover Strait, whereas in reaches upstream (i.e. north) of the tectonic hinge line, in the North Sea basin tectonic depocenter (Figs. 7.5, 7.7 & 7.9), subsidence rates and sediment delivery control gradients (see also Hijma et al. 2012). This affects the pacing of transgressions into these valleys, and thus the preservation potential of Upper Paleolithic and Mesolithic archaeology. As this is known to occur abundantly below the coastal plain and, though more patchily in preservation, also in the nearshore, it is expected also to be present in the offshore Rhine-Meuse and Thames LGM-inherited valley systems (Fig. 7.9b).

Since transgression also occurred for sub-periods within the Early Glacial in this area (towards the MIS 5c and MIS 5a highstands, sea-level curve of Fig. 7.9a), positive taphonomic effects may apply to depositional surfaces with potential Neanderthal sites that occur within the Brown Bank Formation (in the area where this unit preserved, Fig. 7.9b). Tizzard et al. (2014) identified such an Early Glacial transgressive valley fill unit (their Unit 4), and a related unit (Unit 5) that overlies the flanking platform area from which 33 Middle Paleolithic handaxes were dredged. Where these handaxes are attributed a (late) MIS 7 age (Tizzard et al. 2014; Unit 3b), the depositional architecture and taphonomy suggest a protective role for MIS 5c/a transgressive deposits for the platform created at the MIS 7/6 transition, to explain the preserved situation today. This may apply to other areas off East Anglia and in the Thames Estuary (e.g. Dix & Sturt 2011).

Figure 7.7 Scenario maps for paleogeographies of the southern North Sea in the Middle Pleistocene. (A) Interglacial configuration for highstands between 1 Ma and 0.5 Ma (i.e. Cromerian Complex, Fig. 7.4); (B) Interglacial configuration for highstands between ca. 0.42 Ma and 0.17 Ma (i.e. MIS 11, MIS 9, MIS 7; Fig. 7.4). Hijma et al. (2012).

Figure 7.8 Presumed proglacial lake extents in the southern North Sea, (A) within MIS 12, and; (B) within MIS 6. From Gibbard and Cohen (2015). Based on Gibbard (1995; 2007), Busschers et al. (2008) and Moreau et al. (2012). See also Murton and Murton (2012).

Figure 7.9 Reconstructed paleogeography of the southern North Sea in the Late Pleistocene, with a tentative sea-level curve for this timeframe and study area. (A) Last Interglacial highstand and onset of sea-level fall in MIS 5 (ca. 130–80 ka); (B) Last Glacial lowstand of MIS 4 to MIS 2 (ca. 80–20 ka). Hijma et al. (2012).

Figure 7.10 GIA-modeled transgression of the North Sea during the Early Holocene. The Dogger Bank transforms into an island and rapid marine incursion occurs from both the north and the south. Figure created using supplementary images from Sturt et al. (2013), with permission from Elsevier.

Figure 7.11 Time-depth distribution and local mean high water (MHW) error band of sea-level index points obtained from basal peat data from the German Bight (‘Elbe paleovalley’) and Southern Bight (‘Dutch North Sea west’), as compiled from various data sources. Inset shows breakdown by subregion. Early Holocene paleoshorelines of the same age in the Elbe area occur at greater depth than in the Southern Bight, reflecting differential GIA. Figure from Vink et al. (2007).

Figure 7.12 Early Holocene paleolandscape reconstruction derived from 3D seismic data showing the area around the Dogger Bank, spanning the UK and Dutch sectors. Features in blue are mapped water bodies (rivers, estuaries and lakes); color coding shows relative topography. Van Heteren et al. (2014). Reproduced with permission from Cambridge University Press.

Mapping archaeological potential

Since key tectonic and sedimentary controls (i.e. drainage and coastline configurations) appear to have remained largely stable prior to MIS 12 (see Fig. 7.4), a larger part of the early Middle Pleistocene landscape and associated archaeology may have been preserved within the subsiding depocenter off the northwestern Netherlands (at the top of the Yarmouth Roads Formation), and at equivalent depth onshore in the Netherlands. However, this will be deeply buried and hard to access, particularly in the central part of the basin. An exception may be the areas where ice-pushed ridges have been built by displacing blocks of Yarmouth Roads Formation, for instance, at the limit of the MIS 6 glaciation (Moreau et al. 2012). Onshore in the Netherlands, it is from quarry exposures in such Saalian ice-pushed ridges that Middle Paleolithic archaeology (e.g. Rhenen industry), is accessible from formations that otherwise occur tens of meters below the modern coastal plain surface (Stapert 1987).

After MIS 12, the preserved record becomes oriented (aligned) to drainage networks (periglacial fluvial and proglacial fluvial), whereas the main patch of record preserved from the Cromerian Complex is a fragmented coastal plain. From the younger period, the Holocene would hold preserved equivalents of such a coastal plain situation (Early Holocene: offshore; Middle and Late Holocene: onshore). In some places we would expect comparable Last Interglacial deposits (onshore) and Early Glacial deposits (offshore: regressive, interrupted by transgressions, see Fig. 7.9a). From the remaining periods it is expected that mainly mixed-age lag deposits characterize the North Sea floor (Hijma et al. 2012).

Across much of the basin, fluvial systems/ paleochannels, owing to their ubiquity in the geomorphological record and their ability to preserve through aggradation, infilling or floodplain deposition may hold evidence that ranges in its taphonomic condition from in situ to reworked (Hosfield 2007; Ward & Larcombe 2008; Cohen et al. 2012). Superficial marine reworking will be widespread, deep marine reworking by tidal channels localized, and Holocene reworking will often affect deposits that had also experienced Late Pleistocene fluvial (periglacial rivers, Fig. 7.9b) and marine (Last Interglacial and Early Glacial, Fig. 7.9a) reworking cycles.

Attempts have been made to use understanding of these processes combined with offshore geological and geomorphological evidence to indicate or predict where archaeological deposits are preserved. Goodwyn et al. (2010), working at the level of the UK continental shelf, based predictions on paleogeographic reconstructions coupled with evidence of known paleolandscape features (e.g. paleochannels, submerged forests). Simply put, high-potential areas were sub-aerially exposed during lowstands and have known evidence of paleolandscapes. Low-potential areas were never exposed, or were ice-covered, and presently comprise bedrock overlain by seabed sediment; for instance, the erosional Southern Bight. This approach does go some way to addressing shelf-scale taphonomy, but when considering an individual sea basin, suffers from a lack of geological detail and chronological depth. The input sea-level reconstructions are solely based on the post-LGM (Brooks et al. 2011) and given the scale of analysis, little consideration is given to the aforementioned differences in glaciation, sea level, sedimentation, change of drainage systems, GIA and tectonics which result in preservation varying in both space and time.

A more regional-scale conceptual approach, developed by Ward and Larcombe (2008), identified the taphonomic processes at play in particular landscape settings/environments and accordingly assigned a preservation rating to each. For example, floodplains were assigned a high rating because burial by fine-grained flood deposits or within peat-infilling depressions (i.e. channels and basins) was considered to promote preservation of archaeological and paleoenvironmental remains. Conversely, open sandy coasts were regarded as low potential owing to significant reworking by waves (see Ward & Larcombe 2008: table 1 for full list of landscape settings and ratings). In conjunction with information on post-LGM paleogeographical change, Ward and Larcombe (2008) applied these ratings to selected areas of the central and southern North Sea: the Dogger, Brown and Norfolk banks. This accordingly identified areas of high potential around and under these features predominantly where artifacts or fossil bones had previously been found, or where Pleistocene and Holocene peat, fluvial, estuarine or floodplain deposits had previously been identified. However, the absence of detailed geological information prevented spatial localization of areas (or depths) of potential beyond generalized zones (e.g. east of the Brown Bank) or geological formations (e.g. the Elbow Formation).

Similarly, Flemming (2002; 2004) discussed particular landscape settings (e.g. estuaries and river valleys, peat layers, former archipelagos) in which preservation was more/less likely. As in Ward and Larcombe (2008), the lack of high-resolution geological data hindered the assignment of potential beyond large-scale or wide-ranging classifications, such as that the large areas of the central North Sea around the Dogger and Brown banks were of high potential (Flemming 2002). Gaffney et al. (2007; 2009) and Fitch (2011) on the other hand, attempted to map archaeological potential by combining assigned archaeological potential (based on the likelihood of occupation and preservation of a given landscape feature; a similar approach to Ward & Larcombe 2008) with paleolandscape maps derived from 3D seismic data. This mapping identified zones of higher preservation potential centered on the shores of the Outer Silver Pit and around a series of former channels running north of it (see for example Fitch et al. 2007: fig. 9.7). A drawback to this is that it is presently restricted to the Early Holocene landscapes around the Dogger Bank and Outer Silver Pit in the UK sector. Also, extensive archaeological and geological sampling, which could test the model and refine the paleolandscape reconstruction, remains to be done.

Nevertheless, the use of seismic data in conjunction with geological data from cores or boreholes has demonstrably allowed progression from shelf-scale and conceptual assessments of preservation potential. Effectively, detailed geological mapping, particularly if combined with accurate chronological information, affords the possibility to identify the elements of the paleolandscape which have been preserved in a given area and also equally importantly, identify what has not been preserved. Accuracy is of course contingent on the survey and sampling density and, crucially, the availability of secure chronological information (see van Heteren et al. 2014). Such studies are presently limited to small regions within the basin; for instance, off East Anglia (Tizzard et al. 2014) and the Netherlands (Kuitems et al. 2015). The single example that has attempted to go (successfully) from geological mapping to archaeological assessment and sampling is the discovery of the Yangtzehaven site in the Rotterdam Maasvlakte harbor extension (see ‘Evidence of submerged landscapes on the shelf’ pages [xxx]–[xxx]). Part of the secret of the success may be the presence of transgressive deposits protecting the surfaces hosting the archaeology. This suggests that mapping of valley networks with geophysics (swath bathymetry, shallow seismics) and coring the uppermost meters of the seabed to identify whether transgressive units are present between ‘terrestrial substrate’ and ‘active seabed’ may be a useful and informative exercise. This transgressive unit need not be a basal peat everywhere but may well be muds, similar to the floodplains in the Ward and Lacombe (2008) model. While examples for this are mostly drawn from the Dutch and British sectors of the North Sea, this would also hold for the Belgian sector (paleovalleys of the rivers Scheldt and IJzer) and for the German and Danish sectors in the German Bight (major paleovalleys of the Elbe and Weser, and those of their tributaries).

An important issue to consider is the extent to which occasional success in local projects (e.g. a single gravel extraction area or selected construction site) can be utilized at regional-scale (e.g. a wind-farm planning zone) or the scale of entire sectors of the North Sea. Similarly, evaluating to what degree success is due to the relative vicinity to coastal land (with known finds and established taphonomy) matters if the suitability of methodologies for areas further offshore is to be assessed. The drowned landscapes and archaeology of the nearshore and offshore may well have much in common even in taphonomy, but not with respect to amounts of useful prior information.

The level of mapping and sampling of the greater part of the North Sea is insufficient to go beyond relatively broad predictions of archaeological preservation potential (Bicket 2013). Nonetheless, the research done to date shows the potential of what can be achieved with sufficient resources. At the time of writing, work is underway to develop an offshore version of the Netherlands’ Indicative Map of Archaeological Values (IKAW) for the Dutch sector, using as its basis the geological mapping described above (Erkens et al. 2014; Ward et al. 2014). This map identifies zones with higher or lower potential for intact Upper Paleolithic and Mesolithic landscapes. However, when destructive activities such as aggregate mining are planned, additional site-specific research will always be necessary because the current data density does not allow for detailed prediction. Procedures and experience from combined geotechnical, geological and archaeological surveying in future wind-farm areas (e.g. Cotterill et al. 2015) are another form of present research that could be utilized in other parts of the North Sea in the future.

Conclusion

The North Sea basin is an important region for submerged landscape research. It potentially holds some of the earliest evidence for hominin occupation of northwestern Europe and figures strongly in addressing questions on hominin migration and responses to environmental change through the last 500,000 years. From the Early Holocene, it contains the Doggerland area which is known to have been transgressed while occupied by Mesolithic people and forms a contact zone between northerly and central European Mesolithic populations. Importantly, there is confirmation that archaeological, faunal, paleoenvironmental and paleolandscape evidence has been preserved on the continental shelf. However, the evidence base is fragmentary, temporally discontinuous and often chronologically mixed — unsurprising given the Quaternary climatic, paleoenvironmental and paleogeographic changes that have affected the study area and their influence on taphonomic processes. Even so, the extant evidence is a powerful indication that the North Sea was habitable, almost certainly a focus of settlement and dispersal and, crucially, can be effectively studied.

The oceanographic conditions and size of the modern North Sea, coupled with the discontinuous and often buried nature of its Quaternary depositional record, mean that geological approaches to submerged landscape reconstruction and archaeological inventorying are essential (Bicket 2013; Ward et al. 2014; Vos et al. 2015; this chapter). In turn, the effectiveness of these approaches and the level to which the submerged landscape and its contents can be effectively resolved depend on two things: firstly, the taphonomic processes which have either helped preserve or destroy large parts of the evidence; and secondly, the density of geological sampling and geophysical survey. Certain regions will have better preservation than others, and are more suitable for certain sampling and survey techniques than others. It is important to lay out surveying and sampling campaigns in such a way that they address the questions of taphonomy besides more basic lithological characterization of the sub-surface. Hereto pages 170–174 summarized taphonomic variables and their use in mapping archaeological potential at local, regional and super-regional scale. Sections on ‘Quaternary background and paleogeographic framework’ and ‘Evidence of submerged landscapes on the shelf’ (see pages 152–169) supported that assessment by providing an overview of geological history and landform inventory, highlighting links with taphonomy throughout the text.

Recent decades have seen considerable advances relevant to submerged landscape research, in terms of data quantities, technical developments in geological and archaeological surveying and conceptual developments in dealing with the data at new resolutions and integrating it with knowledge acquired from the surrounding onshore regions. Conceptual approaches are now well established (e.g. Flemming 2004; Bailey & Flemming 2008) and research and management frameworks covering the North Sea have been developed (e.g. Peeters et al. 2009; Ward et al. 2014). Paleolandscape reconstructions have also become more accurate and detailed (e.g. compare Coles 1998 with Gaffney et al. 2007; 2009; Hijma et al. 2012), and now afford the possibility to go beyond the shelf-scale geological reports and papers that were for many years state of the art (e.g. Jelgersma 1979; Zagwijn 1979; Cameron et al. 1992; Bridgland & D'Olier 1995; Funnel 1996).

Together with improved reconstruction comes better understanding of taphonomic processes and hence of the preservation potential of archaeological material. However, detailed studies are still the exception rather than the rule and issues remain outstanding regarding chronology and correlation. Furthermore, the ability to take the geological mapping to the level where it can pinpoint archaeological deposits is still in its infancy and in the North Sea to date has only been demonstrated in a few instances.

A further step up in the quantity and quality of data is required to make such techniques work further offshore (see ‘Mapping archaeological potential’, pages 173–174). The research done to date on the submerged landscape of the North Sea provides a secure foundation on which future work can build and also provides examples that can point the way for other less well-studied shelves. Insight into geological developments governing formation and preservation of the Quaternary record and its taphonomy allows regionalization of what archaeology can be expected to be encountered where (this chapter, see also Bicket 2013), aiding landscape-based management of submerged archaeological heritage.

Wind-farm planning activities in the Dogger Bank and Outer Silver Pit areas in the UK sectors may be the next activity to reveal offshore North Sea Mesolithic discoveries. This Mesolithic potential should extend into the Elbe paleovalley in the German Bight and the Danish sector south-west of the Skagerrak. In this region it connects to the rich submerged Mesolithic archaeology in the straits connecting the Skagerrak with the Baltic Sea (e.g. Fischer 1995; Pedersen et al. 1997).

Regarding Middle Paleolithic and older archaeology, aggregate extraction in the UK sector of the southern North Sea is expected to continue to provide occasional discoveries and surveying opportunities (such as Tizzard et al. 2014). Late Middle Paleolithic archaeology (Neanderthals) and associated paleontology is expected to be found in the Belgian sector and the south-west of the Dutch sector, in the offshore continuations of the Last Interglacial Rhine and Meuse valleys. From before the Holocene, what could or should be present between the Dogger Bank, East Anglia, Rhine and Elbe in the central North Sea is the greatest unknown.

Data sources/Useful links

Belgium

• SeArch Project: www.sea-arch.be/en
• Flemish Hydrography: www.vlaamsehydrografie.be
• Flanders Marine Institute (VLIZ): www.vliz.be/en

Denmark

• Geological Survey of Denmark and Greenland (GEUS): www.geus.dk/geuspage-uk.htm
• Danish Geodata Agency: eng.gst.dk

Germany

• Federal Maritime & Hydrographic Agency (BSH): www.bsh.de/de/index.jsp
• Federal Institute for Geosciences & Natural Resources (BGR): www.bgr.bund.de

Norway

• Norwegian Geological Survey (NGU): www.ngu.no/en/node
• Norwegian Mapping Authority (Kartverkert): www.kartverket.no/en/

The Netherlands

• TNO Geological Survey of the Netherlands: www.dinoloket.nl/en
• Royal Netherland Navy Hydrographic Service: www.defensie.nl/english/topics/hydrography
• RCE Cultural Heritage Agency (Rijksdienst Cult. Erfgoed): www.culturalheritageagency.nl/en
• Maasvlakte 2 program: www.20metersunderwater.nl
• Netherlands Enterprise Agency (public datasets on windfarm zones): offshorewind.rvo.nl
• Facebook public outreach site displaying numerous photos of beach finds of bones from the artificial beach of the Rotterdam harbor extension: www.facebook.com/maasvlakte.strandvondsten

UK

• British Geological Survey (BGS): www.bgs.ac.uk
• UK Hydrographic Office (UKHO): www.gov.uk/ government/organisations/uk-hydrographic-office
• British Oceanographic Data Centre: www.bodc.ac.uk
• The Crown Estate: www.thecrownestate.co.uk
• Historic England / English Heritage: www.historicengland.org.uk www.english-heritage.org.uk

General

• EMODnet initiative: www.emodnet.eu
• INQUA Commission Marine Processes: www.inqua. org/cmp/index.html

Acknowledgments

Our colleague Dr. Sytze van Heteren (TNO Geological survey of the Netherlands) is thanked for his constructive comments and suggestions on an earlier draft of this chapter. Editors Nicholas C. Flemming and Anthony Burgess are thanked for their guidance, support and patience throughout the manuscript production.

References

1. Alappat, L., Vink, A., Tsukamoto, S. & Frechen, M. 2010. Establishing the Late Pleistocene–Holocene sedimentation boundary in the southern North Sea using OSL dating of shallow continental shelf sediments. Proceedings of the Geologists’ Association 121:43-54.
2. Allen, J. R. L. 2000. Morphodynamics of Holocene salt marshes: a review sketch from the Atlantic and Southern North Sea coasts of Europe. Quaternary Science Reviews 19:1155-1231.
3. Amkreutz, L. W. S. W., 2013. Persistent Traditions. A Long-term Perspective on Communities in the Process of Neolithisation in the Lower Rhine Area (5500-2500 cal BC). Ph.D thesis. University of Leiden. Sidestone Press Dissertations: Leiden.
4. Anthony, D. & Leth, J. O. 2002. Large-scale bedforms, sediment distribution and sand mobility in the eastern North Sea off the Danish west coast. Marine Geology 183:247-263.
5. Antoine, P., Rousseau, D-. D., Moine, O. et al. 2009. Rapid and cyclic aeolian deposition during the Last Glacial in European loess: a high-resolution record from Nussloch, Germany. Quaternary Science Reviews 28:2955-2973.
6. Arends, F. 1833. Physische Geschichte der Nordsee-Küste und deren Veränderungen durch Sturmfluthen seit der Cymbrischen Fluth bis jetzt. H. Woortman: Emden.
7. Ashton, N., Lewis, S. G., De Groote, I. et al. 2014. Hominin footprints from Early Pleistocene deposits at Happisburgh, UK. PLoS ONE 9(2): e88329. doi:10.1371/journal.pone.0088329.
8. Bailey, G. N. & Flemming, N. C. 2008. Archaeology of the continental shelf: Marine resources, submerged landscapes and underwater archaeology. Quaternary Science Reviews 27:2153-2165.
9. Balson, P., Butcher, A., Holmes, R. et al. 2002. North Sea Geology. Technical report Produced for Strategic Environmental Assessment — SEA2 & SEA3. UK Department of Trade and Industry: London.
10. Barne, J. H, Robson, C. F., Kaznowksa, S. S., Doody, J. P. & Davidson, N. C. (eds.) 1995a. Coasts and Seas of the United Kingdom. Region 5 North-east England: Berwick-upon-Tweed to Filey Bay. Joint Nature Conservation Committee: Peterborough.
11. Barne, J. H, Robson, C. F., Kaznowksa, S. S., Doody, J. P. & Davidson, N. C. (eds.) 1995b. Coasts and Seas of the United Kingdom. Region 6: Eastern England: Flamborough Head to Great Yarmouth. Joint Nature Conservation Committee: Peterborough.
12. Barne, J. H., Robson, C. F., Kaznowksa, S. S. & Doody, J. P. (eds.) 1997. Coasts and Seas of the United Kingdom. Region 4 South-east Scotland: Montrose to Eyemouth. Joint Nature Conservation Committee: Peterborough.
13. Barne, J. H, Robson, C. F., Kaznowksa, S. S., Doody, J. P., Davidson, N. C. & Buck, A.L. (eds.) 1998. Coasts and Seas of the United Kingdom. Region 7 South-east England: Lowestoft to Dungeness. Joint Nature Conservation Committee: Peterborough.
14. Bates, M. R., Bates, C. R. & Briant, R. M. 2007. Bridging the gap: a terrestrial view of shallow marine sequences and the importance of the transition zone. Journal of Archaeological Science 34:1537-1551.
15. Beets, D. J. & van der Spek, A. J. F. 2000. The Holocene evolution of the barrier and the back-barrier basins of Belgium and the Netherlands as a function of late Weichselian morphology, relative sea-level rise and sediment supply. Netherlands Journal of Geosciences 79:3-16.
16. Behre, K.-E. 2004. Coastal development, sea-level change and settlement history during the later Holocene in the Clay District of Lower Saxony (Niedersachsen), northern Germany. Quaternary International 112:37-53.
17. Bell, M., Warren, G., Cobb, H. et al. 2013. The Mesolithic. In Ransley, J. Sturt, F., Dix, J. Adams, J. & Blue, L. (eds.) People and the Sea: a Maritime Archaeological Research Agenda for England. CBA Research Report 171. pp. 30-49. Council for British Archaeology: York.
18. Bicket, A. 2013. Audit of Current State of Knowledge of Submerged Palaeolandscapes and Sites. English Heritage Project No. 6231. Wessex Archaeology: Salisbury.
19. Borst, W., Weerts, H., Vellinga, T., & Otte, A. 2014. Monitoring programme for MV2, Part IV — archaeological and palaeontological finds. Terra et Aqua 135:5-16.
20. Böse, M., Luthgens, C., Lee, J. R. & Rose, J. 2012. Quaternary glaciations of northern Europe. Quaternary Science Reviews 44:1-25.
21. Bridgland, D. R., & D'Olier, B. 1995. The Pleistocene evolution of the Thames and Rhine drainage systems in the southern North Sea Basin. Geological Society London, Special Publications 96:27-45.
22. Bridgland, D. R., D'Olier, B., Gibbard, P. L. & Roe, H. M. 1993. Correlation of Thames terrace deposits between the Lower Thames, eastern Essex and the submerged offshore continuation of the Thames-Medway valley. Proceedings of the Geologists Association 104:51-57.
23. Briggs, K., Thomson, K. & Gaffney, V. 2007. A geomorphological Investigation of submerged depositional features within the Outer Silver Pit, Southern North Sea. In Gaffney V., Thomson, K. & Fitch, S. (eds.) 2007. Mapping Doggerland: The Mesolithic Landscapes of the Southern North Sea. pp. 43-59. Archaeopress: Oxford.
24. Brooks, A. J., Bradley, S. L., Edwards, R. J. & Goodwyn, N. 2011. The palaeogeography of Northwest Europe during the last 20,000 years. Journal of Maps 7:573-587.
25. Busschers, F. S., Kasse, C., van Balen, R. T. et al. 2007. Late Pleistocene evolution of the Rhine-Meuse system in the southern North Sea basin: imprints of climate change, sea-level oscillation and glacio-isostacy. Quaternary Science Reviews 26:3216-3248.
26. Busschers, F. S., van Balen, R. T., Cohen, K. M. et al. 2008. Response of the Rhine-Meuse fluvial system to Saalian ice-sheet dynamics. Boreas 37:377-398.
27. Busschers, F. S., Wesselingh, F., Kars, R. H. et al. 2010. Radiocarbon dating of Late Pleistocene marine shells from the southern North Sea. Radiocarbon 56:1151-1166.
28. Cameron, T. D., Crosby, A., Balson, P. S. et al. 1992. UK Offshore Regional report: The Geology of the Southern North Sea. HMSO for the British Geological Survey: London.
29. Carr, S. J. 2004. The North Sea basin. In Ehlers, J. & Gibbard, P. L. (eds.) Quaternary Glaciations — Extent and Chronology. Part 1 Europe. pp. 261-270. Elsevier: Amsterdam.
30. Carr, S. J., Holmes, R., van der Meer, J.M. & Rose, J. 2006. The Last Glacial Maximum in the North Sea Basin: micromorphological evidence of extensive glaciation. Journal of Quaternary Science 21:131-153.
31. Chang, T. S., Flemming, B. W., Tilch, E., Bartholomä, A. & Wöstmann, R. 2006. Late Holocene stratigraphic evolution of a back-barrier tidal basin in the East Frisian Wadden Sea, southern North Sea: transgressive deposition and its preservation potential. Facies 52: 329-340.
32. Clark, C. D., Hughes, A. L. C., Greenwood, S. L., Jordan, C. & Sejrup, H. P. 2012. Pattern and timing of retreat of the last British-Irish Ice Sheet. Quaternary Science Reviews 44:112-146.
33. Cloetingh, S., Ziegler, P. A., Beekman, F. et al. 2005. Lithospheric memory, state of stress and rheology: neotectonic controls on Europe's intraplate continental topography. Quaternary Science Reviews 24:241-304.
34. Cohen, K. M. & Lobo, F. J. 2013. Continental shelf drowned landscapes: Submerged geomorphological and sedimentary record of the youngest cycles (special issue editorial). Geomorphology 203:1-5.
35. Cohen, K. M., MacDonald, K., Joordens, J. C. A, Roebroeks, W. & Gibbard, P. L. 2012. The earliest occupation of north-west Europe: a coastal perspective. Quaternary International 271:70-83.
36. Cohen, K. M., Gibbard, P. L. & Weerts, H. J. T. 2014. North Sea palaeogeographical reconstructions for the last 1 Ma. Netherlands Journal of Geosciences 93:7-29.
37. Coles, B. J. 1998. Doggerland: a speculative survey. Proceedings of the Prehistoric Society 64:45-81.
38. Cotterill, C., James, L., Long, D., Forsberg, C-. F., Tjelta, T. I. & Mulley S. 2015. Dogger Bank and the case of the complex stratigraphy (abstract). Quaternary Research Association Annual Discussion Meeting. 6^th–8^th January 2015, Edinburgh, p. 1.
39. Dawson, A. G., Smith, D. E. & Long, D. 1990. Evidence for a tsunami from a Mesolithic site in Inverness, Scotland. Journal of Archaeological Science 17:509-512.
40. Devoy, R. J. N. 1979. Flandrian sea level changes and vegetational history of the lower Thames Estuary. Philosophical Transactions of the Royal Society of London B: Biological Sciences 285:355-407.
41. Dix, J. K & Sturt, F. C. 2011. The Relic Palaeo-landscapes of the Thames Estuary. Report prepared for the Marine Aggregate Levy Sustainability Fund.
42. Dong, P. & Guzzetti, F. 2005. Frequency-size statistics of coastal soft-cliff erosion. Journal of Waterway, Port, Coastal and Ocean Engineering 131:37-42.
43. Ehlers, J. & Gibbard, P. L. (eds.) 2004. Quaternary Glaciations - Extent and Chronology. Part I: Europe. Elsevier: Amsterdam.
44. EMODnet 2014. Seabed substrate map. EMODnet Geology website: www.emodnet-geology.eu/.
45. Erkens, G., Hijma, M. P., Peeters, J., van Heteren, S., Marges, V. & Vonhögen-Peeters, L. 2014. Proef Indicatieve Kaart Archeologische Waarden (IKAW) Noordzee. Deltares Report 1206731.
46. Figge, K. 1980. Das Elbe-Urstromtal im Bereich der Deutschen Bucht (Nordsee). Eiszeitalter und Gegenwart 30: 203-211.
47. Fischer, A. 1995. An entrance to the Mesolithic world below the ocean. Status of ten years’ work on the Danish sea floor. In Fischer, A. (ed.) Man and Sea in the Mesolithic. pp. 371-384. Oxbow Books: Oxford.
48. Fitch, S. E. J. 2011. The Mesolithic landscape of the southern North Sea. Ph.D Thesis. University of Birmingham, UK.
49. Fitch, S., Thomson, K., & Gaffney, V. L. 2005. Late Pleistocene and Holocene depositional systems and the palaeogeography of the Dogger Bank, North Sea. Quaternary Research 64:185-196.
50. Fitch, S., Gaffney, V. L. & Thomson, K. 2007. The archaeology of North Sea palaeolandscapes. In Gaffney, V., Thomson, K. & Fitch, S. (eds.) Mapping Doggerland: The Mesolithic Landscapes of the Southern North Sea. pp. 105-118. Archaeopress: Oxford.
51. Flemming, N. C. 2002. The Scope of Strategic Environmental Assessment of North Sea areas SEA3 and SEA2 in Regard to Prehistoric Archaeological Remains. UK Department of Trade and Industry: London.
52. Flemming, N. C. (ed.) 2004. Submarine Prehistoric Archaeology of the North Sea. CBA Research Report 141. Council of British Archaeology: York.
53. Fruergaard, M., Andersen, T. J., Nielsen, L. H., Johannessen, P. N., Aagaard, T. & Pejrup, M. 2015. High-resolution reconstruction of a coastal barrier system: impact of Holocene sea-level change. Sedimentology 62:928-969.
54. Funnell, B. M. 1995. Global sea-level and the (pen-)insularity of late Cenozoic Britain. In Preece, R. C. (ed.) Island Britain: a Quaternary Perspective. pp. 3-13. Geological Society: London.
55. Funnel, B. M. 1996. Plio-Pleistocene palaeogeography of the southern North Sea Basin (3.75-0.60 Ma). Quaternary Science Reviews 15:391-405.
56. Gaffney, V., Thomson, K. & Fitch, S. (eds.) 2007. Mapping Doggerland: The Mesolithic Landscapes of the Southern North Sea. Archaeopress: Oxford.
57. Gaffney V., Fitch, S. & Smith, D. 2009. Europe's Lost World: the rediscovery of Doggerland. CBA Research Report 160. Council for British Archaeology: York.
58. Gatliff, P. W., Richards, P. C., Smith, K. et al. 1994. UK Offshore Regional Report: The Geology of the Central North Sea. HMSO for the British Geological Survey: London.
59. Gibbard, P. L. 1988. The history of the great northwest European rivers during the past three million years. Philosophical Transactions of the Royal Society, London B318:559-602.
60. Gibbard, P. L. 1995. The formation of the Strait of Dover. In Preece, R. C. (ed.) Island Britain: a Quaternary Perspective. pp. 15-26. Geological Society: London.
61. Gibbard, P. L. 2007. Palaeogeography: Europe cut adrift. Nature 448:259-260.
62. Gibbard, P. L. & Cohen, K. M. 2008. Global chronostratigraphical correlation table for the last 2.7 million years. Episodes 31:243-247.
63. Gibbard, P. L. & Cohen, K. M. 2015. Quaternary evolution of the North Sea and the English Channel. Proceedings of the Open University Geological Society 1:65-75.
64. Gibbard, P. L., Turner, C. & West, R. G. 2013. The Bytham river reconsidered. Quaternary International 292:15-32.
65. Glimmerveen, J., Mol, D., Post, K. et al. 2004. The North Sea project: the first palaeontological, palynological and archaeological results. In Flemming, N. C. (ed.) Submarine Prehistoric Archaeology of the North Sea. CBA Research Report 141. pp. 43-52. Council of British Archaeology: York.
66. Glimmerveen, J., Mol, D. & van der Plicht, H. 2006. The Pleistocene reindeer of the North Sea — initial palaeontological data and archaeological remarks. Quaternary International 142-143:242-246.
67. Goodwyn, N., Brooks, A.J. & Tillin, H. 2010. Waterlands: Developing Management Indicators for Submerged Palaeoenvironmental Landscapes. Report prepared under the Marine Aggregates Levy Sustainability Fund (Ref No. MEPF 09/P109). ABP Marine Environmental Research.
68. Gupta, S., Collier, J. S., Palmer-Felgate, A. & Potter, G. 2007. Catastrophic flooding origin of shelf valley systems in the English Channel. Nature 448:342-345.
69. Gyllencreutz, R. 2005. Late Glacial and Holocene paleoceanography in the Skagerrak from high-resolution grain size records. Palaeogeography, Palaeoclimatology, Palaeoecology 222:344-369.
70. Hazell, Z. J. 2008. Offshore and intertidal peat deposits, England — a resource assessment and development of a database. Environmental Archaeology 13:101-110.
71. Hijma, M. P., & Cohen, K. M. 2010. Timing and magnitude of the sea-level jump preluding the 8200 yr event. Geology 38:275-278.
72. Hijma, M. P., & Cohen, K. M. 2011. Holocene transgression of the Rhine river mouth area, The Netherlands/Southern North Sea: palaeogeography and sequence stratigraphy. Sedimentology 58:1453-1485.
73. Hijma, M. P., van der Spek, A. J. F. & van Heteren, S. 2010. Development of a mid-Holocene estuarine basin, Rhine-Meuse mouth area, offshore The Netherlands. Marine Geology 271:198-211.
74. Hijma, M. P., Cohen, K. M., Roebroeks, W., Westerhoff, W. E. & Busschers, F. S. 2012. Pleistocene Rhine-Thames landscapes: geological background for hominin occupation of the southern North Sea region. Journal of Quaternary Science 27:17-39.
75. Hijma, M. P., Engelhart, S. E., Törnqvist, T. E., Horton, B. P, Hu, P. & Hill, D. F. 2015. A protocol for a geological sea-level database. In Shennan, I., Long, A. J. & Horton, B. P. (eds.) Handbook of Sea-Level Research. pp. 536-553. John Wiley & Sons: Chichester.
76. Hosfield, R. T. 2007. Terrestrial implications for the maritime geoarchaeological resource: A view from the Lower Palaeolithic. Journal of Maritime Archaeology 2:4-23.
77. Houmark-Nielsen, M. & Kjær, K. H. 2003. Southwest Scandinavia, 40–15 kyr BP: palaeogeography and environmental change. Journal of Quaternary Science 18:769-786.
78. Housley, R. A. 1991. AMS dates from the Late Glacial and early Postglacial in North-west Europe: A review. In N. Barton, Roberts A. J. & Roe, D. A (eds.) The Late Glacial in North-West Europe. Human Adaption and Environmental Change at the End of the Pleistocene. CBA Research Report 77. pp. 25-39. Council for British Archaeology: London.
79. Hublin, J-. J., Weston, D., Gunz, P. et al. 2009. Out of the North Sea: the Zeeland Ridges Neandertal. Journal of Human Evolution 57:777-785.
80. Huthnance, J. M. 1991. Physical oceanography of the North Sea. Ocean & Shoreline Management 16:199-231.
81. Huuse, M. & Lykke-Andersen, H. 2000. Overdeepened Quaternary valleys in the eastern Danish North Sea: morphology and origin. Quaternary Science Reviews 19:1233-1253.
82. Jelgersma, S. 1979. Sea-level changes in the North Sea Basin. In Oele, E., Schüttenhelm, R. T. E. & Wiggers, A. J. (eds.) The Quaternary History of the North Sea. Acta Universitatis Upsaliensis Symposia Universitatis Upsaliensis Annum Quingentesimum Celebrantis 2:233-248.
83. Konradi, P. B. 2000. Biostratigraphy and environment of the Holocene marine transgression in the Heligoland Channel, North Sea. Bulletin of the Geological Society of Denmark 47:71-79.
84. Kooi, H., Johnston, P., Lambeck, K., Smither, C. & Molendijk, R. 1998. Geological causes of recent (∼100 yr) vertical land movement in the Netherlands. Tectonophysics 299:297-316.
85. Kuitems, M., van Kolfschoten, Th., Busschers, F. & De Loecker, D. 2015. The Geoarchaeological and Palaeontological research in the Maasvlakte 2 sand extraction zone and on the artificially created Maasvlakte 2 beach — a synthesis. In Moree, J. M. & Sier, M. M. (eds.) Interdisciplinary Archaeological Research Programme Maasvlakte 2, Rotterdam. BOORrapporten 566. pp. 351-398. Gemeente Rotterdam / Rijksdienst voor het Cultureel Erfgoed & Port of Rotterdam: Rotterdam.
86. Laban C. 1995. The Pleistocene Glaciations in the Dutch Sector of the North Sea. Ph.D thesis, University of Amsterdam, the Netherlands.
87. Laban, C. & van der Meer, J. J. M. 2011. Pleistocene Glaciation in The Netherlands. In Ehlers, J., Gibbard, P. L. & Hughes P. D. (eds.) Quaternary Glaciations— Extent and Chronology. A Closer Look. Developments in Quaternary Sciences vol. 15. pp. 247-260. Elsevier: Amsterdam.
88. Lambeck, K. 1995. Late Devensian and Holocene shorelines of the British Isles and the North Sea from models of glacio-hydro-isostatic rebound. Journal of the Geological Society 152:437-448.
89. Lambeck, K., Purcell, A., Funder, S., Kjær, K. H., Larsen, E. & Moller, P. 2006. Constraints on the Late Saalian to early Middle Weichselian ice sheet of Eurasia from field data and rebound modelling. Boreas 35:539-575.
90. Lee, J. R., Busschers, F. S. & Sejrup, H. P. 2012. Pre-Weichselian Quaternary glaciations of the British Isles, The Netherlands, Norway and adjacent marine areas south of 68°N: implications for long-term ice sheet development in northern Europe. Quaternary Science Reviews 44:213-228.
91. Lekens, W. A. H., Sejrup, H. P., Haflidason, H., Knies, J., & Richter, T. 2006. Meltwater and ice rafting in the southern Norwegian Sea between 20 and 40 calendar kyr BP: Implications for Fennoscandian Heinrich events. Paleoceanography 21 PA3013 doi:10.1029/2005PA001228.
92. Long, D., Wickham-Jones, C. R. & Ruckley, N. A. 1986. A flint artefact from the northern North Sea. In Roe, D. A. (ed.) Studies in the Upper Palaeolithic of Britain and North West Europe. BAR International Series (No. 296). pp. 55-62. British Archaeological Reports: Oxford.
93. Louwe Kooijmans, L. P. 1975. Benen jacht- en visgerei uit de Midden-Steentijd, gevonden op de Maasvlakte (circa 7000 v.Chr.). In Louwe Kooijmans, L. P. & Sarfatij, H. & Verhoeven, A. (eds.) Archeologen werken in Zuid-Holland, Opgravingen en vondsten uit de laatste 15 jaar. p. 11. Rijksmuseum van Oudheden/Leiden tentoonstellingscatalogus: Leiden.
94. Louwe Kooijmans, L. P. 1980. Archaeology and coastal change in the Netherlands. In Thompson, F. H. (ed.) Archaeology and Coastal Change. pp. 106-133. The Society of Antiquaries of London: London.
95. Louwe Kooijmans, L. P. 2005. Jagerskamp in de moerassen. De donken bij Hardinxveld. In Louwe Kooijmans, L. P., van den Broeke, P. W, Fokkens, H & van Gijn, A. (eds.) Nederland in de prehistorie. pp. 183-186. Bert Bakker: Amsterdam.
96. Lunkka, J. P. 1994. Sedimentation and Iithostratigraphy of the North Sea Drift and Lowestoft Till Formations in the Coastal cliffs of northeast Norfolk, England. Journal of Quaternary Science 9:209-233.
97. Maddy, D., Bridgland, D. R. & Green, C. P. 2000. Crustal uplift in southern England: evidence from the river terrace records. Geomorphology 33:167-181.
98. Martinius, A. W. & van den Berg, J. H. 2011. Atlas of Sedimentary Structures in Estuarine and Tidally-Influenced River Deposits of the Rhine-Meuse-Scheldt System. EAGE Publications bv: Houten.
99. Mathys, M. 2009. The Quaternary Geological Evolution of the Belgian Continental Shelf, Southern North Sea. Ph.D Thesis. Ghent University, Belgium.
100. Meijer, T. & Preece. R. C. 1995. Malacological evidence relating to the insularity of the British Isles during the Quaternary. In Preece, R. (ed.) Island Britain: a Quaternary Perspective. Geological Society London Special Publication 96:89-110.
101. Milne, G. A. 2015. Glacial isostatic adjustment. In Shennan, I., Long, A. J. & Horton, B. P. (eds.) Handbook of Sea-Level Research. pp. 421-438. John Wiley & Sons: Chichester.
102. Mol, D., Post, K., Reumer, J. W. F. et al. 2006. The Eurogeul - first report of the palaeontological, palynological and archaeological investigations of this part of the North Sea. Quaternary International 142-143:178-185.
103. Moreau, J., Huuse, M., Janszen, A., van der Vegt, P., Gibbard, P. L. & Moscariello, A. 2012. The Glaciogenic Unconformity of the Southern North Sea. Geological Society London Special Publications 368:99-110.
104. Moree, J. M. & Sier, M. M. (eds.) 2014. Twintig meter diep! Mesolithicum in de Yangtzehaven — Maasvlakte te Rotterdam. Landschapsontwikkeling en bewoning in het Vroeg Holoceen. BOORrapporten 523. Bureau Oudheidkundig Onderzoek Rotterdam (BOOR): Rotterdam.
105. Moree, J. M. & Sier, M. M. (eds.) 2015. Twenty metres deep! The Mesolithic Period at the Yangtze Harbour site — Rotterdam Maasvlakte, the Netherlands. Early Holocene landscape development and habitation. In Moree, J. M. & Sier, M. M. (eds.) Interdisciplinary Archaeological Research Programme Maasvlakte 2, Rotterdam. BOORrapporten 566. pp. 7-350. Gemeente Rotterdam / Rijksdienst voor het Cultureel Erfgoed & Port of Rotterdam: Rotterdam.
106. Murton, D. K. & Murton, J. B. 2012. Middle and Late Pleistocene glacial lakes of lowland Britain and the southern North Sea Basin. Quaternary International 260:115-142.
107. Neill, S. P., Scourse, J. D. & Uehara, K. 2010. Evolution of bed shear stress distribution over the northwest European shelf during the last 12,000 years. Ocean Dynamics 60:1139-1156.
108. Oost, A. P. 1995. Dynamics and Sedimentary Developments of the Dutch Wadden Sea with a Special Emphasis on the Frisian Inlet: a Study of the Barrier Islands, Ebb-tidal Deltas, Inlets and Drainage Basins. Ph.D Thesis. Utrecht University, the Netherlands.
109. Otto, L., Zimmerman, J. T. F, Furnes, G. K., Mork, M., Sætre, R. & Becker, G. 1990. Review of the physical oceanography of the North Sea. Netherlands Journal of Sea Research 26:161-238.
110. Overeem, I., Weltje, G. J., Bishop-Kay, C. & Kroonenberg, S. B. 2001. The Late Cenozoic Eridanos delta system in the Southern North Sea Basin: a climate signal in sediment supply? Basin Research 13:293-312.
111. Paramor, O. A. L., Allen, K. A., Aanensen, M. et al. 2009. MEFEPO North Sea Atlas. University of Liverpool: Liverpool.
112. Parfitt, S. A., Barendregt, R. W., Breda, M. et al. 2005. The earliest record of human activity in northern Europe. Nature 438:1008-1012.
113. Parfitt, S. A., Ashton, N. M., Lewis, S. G., et al. 2010. Early Pleistocene human occupation at the edge of the boreal zone in northwest Europe. Nature 466:229-233.
114. Pawley, S. M., Bailey, R. M., Rose, J. et al. 2008. Age limits on Middle Pleistocene glacial sediments from OSL dating, north Norfolk, UK. Quaternary Science Reviews 27:1363-1377.
115. Pedersen, J. B. T., Svinth, S. & Bartholdy, J. 2009. Holocene evolution of a drowned melt-water valley in the Danish Wadden Sea. Quaternary Research 72:68-79.
116. Pedersen, L. D., Fischer, A. & Aaby, B. (eds.) 1997. The Danish Storebaelt since the Ice Age: Man, Sea and Forest. Storebaeltsforbindelsen: Copenhagen.
117. Peeters, H., Murphy, P. & Flemming, N. C. (eds.) 2009. North Sea Prehistory Research and Management Framework (NSPRMF) 2009. English Heritage and RCEM: Amersfoort.
118. Peeters, J., Busschers, F. S., & Stouthamer, E. 2015a. Fluvial evolution of the Rhine during the last interglacial\-glacial cycle in the southern North Sea basin: A review and look forward. Quaternary International 357:176-188.
119. Peeters, J. H. M. & Cohen, K. M. 2014. Introduction to North Sea submerged landscapes and prehistory. Netherlands Journal of Geosciences 93:3-5.
120. Peeters, J. H. M. & Momber, G. 2014. The southern North Sea and the human occupation of northwest Europe after the Last Glacial Maximum. Netherlands Journal of Geosciences 93:55-70.
121. Peeters, J. H. M., Brinkhuizen, D. C., Cohen, K. M. et al. 2014. Synthese - Mesolithicum in de Yangtzehaven-Maasvlakte te Rotterdam. In Moree, J. M. & Sier, M. M. (eds.) Twintig meter diep! Mesolithicum in de Yangtzehaven — Maasvlakte te Rotterdam. Landschapsontwikkeling en bewoning in het Vroeg Holoceen. BOORrapporten 523. pp. 287-318. Bureau Oudheidkundig Onderzoek Rotterdam (BOOR): Rotterdam.
122. Peeters, J. H. M., Brinkhuizen, D. C., Cohen, K. M. et al. 2015b. Synthesis — The Mesolithic in the Yangtzeharbour, Maasvlakte, Rotterdam. In Moree, J. M. & Sier, M. M. (eds.) 2015. Twenty metres deep! The Mesolithic Period at the Yangtze Harbour site. BOORrapporten 566 (Interdisciplinary Archaeological Research Programme Maasvlakte 2, Rotterdam). pp. 287-318. Port of Rotterdam Harbour Authority: Rotterdam.
123. Peeters. J., Busschers, F. S., Stouthamer, E. et al. 2016. Sedimentary architecture and chronostratigraphy of a late Quaternary incised-valley fill: A case study of the late Middle and Late Pleistocene Rhine system in the Netherlands. Quaternary Science Reviews 131(A):211-236.
124. Peltier, W. R. 2004. Global glacial isostasy and the surface of the ice-age Earth: The ICE-5G (VM2) model and GRACE. Annual Review of Earth and Planetary Sciences 32:111-149.
125. Peltier, W. R., Shennan, I., Drummond, R. & Horton B. 2002. On the postglacial isostatic adjustment of the British Isles and the shallow viscoelastic structure of the Earth. Geophysical Journal International 148:443-475.
126. Pettitt, P. & White, M. 2012. The British Palaeolithic: Human Societies at the Edge of the Pleistocene World. Routledge: Abingdon/New York.
127. Praeg, D. 2003. Seismic imaging of mid-Pleistocene tunnel-valleys in the North Sea Basin — high resolution from low frequencies. Journal of Applied Geophysics 53:273-298.
128. Reid, C. 1913. Submerged Forests. Cambridge University Press: Cambridge.
129. Rieu, R., van Heteren, S., van der Spek, Ad. J. F. & De Boer, P. L. 2005. Development and preservation of a mid-Holocene tidal-channel network offshore the western Netherlands. Journal of Sedimentary Research 75:409-419.
130. Roe, H. M. & Preece, R. C. 2011. Incised palaeo-channels of the late Middle Pleistocene Thames: age, origins and implications for fluvial palaeogeography and sea-level reconstruction in the southern North Sea basin. Quaternary Science Reviews 30:2498-2519.
131. Roebroeks, W. 2014. Terra incognita: The Palaeolithic record of northwest Europe and the information potential of the southern North Sea. Netherlands Journal of Geosciences 93:43-53.
132. Roep, Th. B., Holst, H., Vissers, R. L. M., Pagnier, H. & Postma, D. 1975. Deposits of southward-flowing, Pleistocene rivers in the Channel region, near Wissant, NW France. Palaeogeography, Palaeoclimatology, Palaeoecology 17:289-308.
133. Rohling, E. J., Grant, K., Bolshaw, M. et al. 2009. Antarctic temperature and global sea level closely coupled over the past five glacial cycles. Nature Geoscience 2:500-504.
134. Rose, J. 1994. Major river systems of central and southern Britain during the Early and Middle Pleistocene. Terra Nova 6:435-443.
135. Rose, J. 2009. Early and Middle Pleistocene landscapes of eastern England. Proceedings of the Geologists' Association 120:3-33.
136. Rose, J., Candy, I., Moorlock, B. S. P. et al. 2002. Early and early Middle Pleistocene river, coastal and neotectonic processes, southeast Norfolk, England. Proceedings of the Geologists' Association 113:47-67.
137. Rydgren, K. & Bondevik, S. 2015. Moss growth patterns and timing of human exposure to a Mesolithic tsunami in the North Atlantic. Geology 43:111-114.
138. SeArch Project 2014. Map showing an overview of the archaeological findings. SeArch website: www.sea-arch.be. Last accessed March 2017.
139. Sejrup, H. P., Larsen, E., Haflidason, H. et al. 2003. Configuration, history and impact of the Norwegian Channel Ice Stream. Boreas 32:18-36.
140. Sejrup, H. P., Nygård, A, Hall A. M. & Haflidason. H. 2009. Middle and Late Weichselian (Devensian) glaciation history of south-western Norway, North Sea and eastern UK. Quaternary Science Reviews 28:370-380.
141. Shennan, I., Lambeck, K., Flather, R. et al. 2000. Modelling western North Sea palaeogeographies and tidal changes during the Holocene. In Shennan, I. & Andrews, J. (eds.) Holocene Land-Ocean Interaction and Environmental Change around the North Sea. Geological Society London Special Publication 166:299-319.
142. Shennan, I., Bradley, S., Milne, G., Brooks, A., Bassett, S. & Hamilton, S. 2006. Relative sea-level changes, glacial isostatic modelling and ice-sheet reconstructions from the British Isles since the Last Glacial Maximum. Journal of Quaternary Science 21:585-599.
143. Shennan, I., Long, A. J. Horton, B. P. (eds.) 2015. Handbook of Sea-Level Research. John Wiley & Sons: Chichester.
144. Sier, M. J., Peeters, J., Dekkers, M. et al. 2015. The Blake Event recorded near the Eemian type locality - A diachronic onset of the Eemian in Europe. Quaternary Geochronology 28:12-28.
145. Smith, A. J. 1985. A catastrophic origin for the palaeovalley system of the eastern English Channel. Marine Geology 64:65-75.
146. Smith, D. E., Harrison, S., Firth, C. R. & Jordan, J. T. 2011. The early Holocene sea level rise. Quaternary Science Reviews, 30:1846-1860.
147. Stapert, D. 1987. A progress report on the Rhenen industry (Central Netherlands) and its stratigraphical context. Palaeohistoria 29:219-243.
148. Steffen, H. & Wu, P. 2011. Glacial isostatic adjustment in Fennoscandia — a review of data and modeling. Journal of Geodynamics 52:169-204.
149. Stewart, M. A., Lornegan, L. & Hampson, G. 2013. 3D seismic analysis of buried tunnel valleys in the central North Sea: morphology, cross-cutting generations and glacial history. Quaternary Science Reviews 72:1-17.
150. Streif, H. 2004. Sedimentary record of Pleistocene and Holocene marine inundations along the North Sea coast of Lower Saxony, Germany. Quaternary International 112:3-28.
151. Stringer, C. 2006. Homo Britannicus: the incredible story of human evolution. Penguin Books: London.
152. Sturt, F. 2006. Local knowledge is required: a rhythmanalytical approach to the late Mesolithic and early Neolithic of the East Anglian Fenland, UK. Journal of Maritime Archaeology 1:119-139.
153. Sturt, F., Garrow, D. & Bradley, S. 2013. New models of North West European Holocene palaeogeography and inundation. Journal of Archaeological Science 40:3963-3976.
154. Teller, J. T., Leverington, D. W. & Mann, J. D. 2002. Freshwater outbursts to the oceans from glacial Lake Agassiz and their role in climate change during the last deglaciation. Quaternary Science Reviews 21:879-887.
155. Tizzard, L., Baggaley, P. A. & Firth, A. J. 2011. Seabed Prehistory: investigating palaeolandsurfaces with Palaeolithic remains from the southern North Sea. In Benjamin, J., Bonsall, C., Pickard, C. & Fischer, A. (eds.) Submerged Prehistory. pp. 65-74. Oxbow: Oxford.
156. Tizzard, L., Bicket, A. R., Benjamin, J. & De Loecker, D. 2014. A Middle Palaeolithic site in the southern North Sea: investigating the archaeology and palaeogeography of Area 240. Journal of Quaternary Science 29:698-710.
157. Törnqvist, T. E., Rosenheim, B. E., Hu, P. & Fernandez, A. B. 2015. Radiocarbon dating and calibration. In Shennan, I., Long, A. J. & Horton, B. P. (eds.) Handbook of Sea-Level Research. pp. 349-360. John Wiley & Sons: Chichester.
158. Toucanne, S., Zaragosi, S., Bourillet, J. F. et al. 2009. Timing of massive ‘Fleuve Manche’ discharges over the last 350 kyr: insights into the European ice-sheet oscillations and the European drainage network from MIS 10 to 2. Quaternary Science Reviews 28:1238-1256.
159. Turney, C. S. M. & Brown, H. 2007. Catastrophic early Holocene sea-level rise, human migration and the Neolithic transition in Europe. Quaternary Science Reviews 26:2036-2041.
160. van den Berg, J. H., Jeuken, M. C. J. L. & Van der Spek, A. J. F. 1996. Hydraulic processes affecting the morphology and evolution of the Westerscheldt Estuary. In Nordstrom, K. F. & Roman, C. T. (eds.) Estuarine Shores: Evolution, Environments and Human Alterations. pp. 157-184. John Wiley & Sons: Chichester.
161. van der Molen, J. 2002. The influence of tides, wind and waves on the net sand transport in the North Sea. Continental Shelf Research 22:2739-2762.
162. van der Plicht, J. & Palstra, S. W. L. 2016. Radiocarbon and mammoth bones: What's in a date. Quaternary International 406 (part B):246-251
163. van der Spek, A. J. F. 1996. Large-scale Evolution of Holocene Tidal Basins in the Netherlands. Ph.D thesis. Utrecht University, the Netherlands.
164. van Geel, B., Bohncke S. J. P. & Dee, H. 1980. A palaeoecological study of an upper Late Glacial and Holocene sequence from ‘De Borchert’, The Netherlands. Review of Palaeobotany and Palynology 31:367-392, 397-448.
165. van Heteren, S., Meekes, J. A. C., Bakker, M. A. J. et al. 2014. Reconstructing North Sea palaeolandscapes from 3D and high-density 2D seismic data: An overview. Netherlands Journal of Geosciences 93:31-42.
166. van Kolfschoten, T. & van Essen, H. 2004. Palaeozoological heritage from the bottom of the North Sea. In Flemming, N. C. (ed.) Submarine Prehistoric Archaeology of the North Sea. CBA Research Report 141. pp. 70-80. Council of British Archaeology: York.
167. van Vliet-Lanoë, B., Laurent, M., Bahain, J. L. et al. 2000. Middle Pleistocene raised beach anomalies in the English Channel: regional and global stratigraphic implications. Journal of Geodynamics 29:15-41.
168. van Vliet-Lanoë, B., Vandenberghe, N., Laurent, M. et al. 2002. Palaeogeographic evolution of northwestern Europe during the Upper Cenozoic. Geodiversitas 24: 511-541.
169. Verhart, L. B. M. 2004. The implications of prehistoric finds on and off the Dutch coast. In Flemming, N. C. (ed.) Submarine Prehistoric Archaeology of the North Sea. CBA Research Report 141. pp. 57-61. Council of British Archaeology: York.
170. Vink, A., Steffen, H., Reinhardt, L. & Kaufmann, G. 2007. Holocene relative sea-level change, isostatic subsidence and the radial viscosity structure of the mantle of northwest Europe (Belgium, the Netherlands, Germany, southern North Sea). Quaternary Science Reviews 26:3249-3275.
171. Vis, G. J., Cohen, K. M., Westerhoff, W. E. et al. 2015. Paleogeography. In Shennan, I., Long, A. J. & Horton, B.P. (eds.) Handbook of Sea-Level Research. pp. 514-535. John Wiley & Sons: Chichester.
172. Vos, P. C. 2015. Compilation of the Holocene palaeogeographical maps of the Netherlands. In Vos, P. C. Origin of the Dutch Coastal Landscape: Long-term Landscape Evolution of the Netherlands During the Holocene, Described and Visualized in National, Regional and Local Palaeogeographical Maps Series. pp. 50-79. Barkhuis Publishing: Deltares / Groningen.
173. Vos, P. C. & van Heeringen R. M. 1997. Holocene geology and occupation history of the province of Zeeland. Mededelingen Nederlands Insitituut voor Toegepaste Geowetenschappen 59:5-109.
174. Vos, P., Bazelmans, J., Weerts, H. & van der Meulen, M. (eds.) 2011. Atlas van Nederland in het Holoceen: Landschap en bewoning vanaf de laatste IJstijd tot nu. Bert Bakker: Amsterdam (in Dutch).
175. Vos, P. C., de Kleine, M. P. C. & Rutten, G. L. 2012. Efficient stepped approach to site investigation for underwater archaeological studies. The Leading Edge 31:940-944.
176. Vos, P. C., Bunnik, F. P. M., Cohen, K. M. & Cremer, H. 2015. A staged geogenetic approach to underwater archaeological prospection in the Port of Rotterdam (Yangtzehaven, Maasvlakte, the Netherlands): A geological and palaeoenvironmental case study for local mapping of Mesolithic lowland landscapes. Quaternary International 367:4-31.
177. Waddington, C., Bailey, G. & Milner, N. 2007. Howick: discussion and interpretation. In Waddington, C. (ed.) Mesolithic Settlement in the North Sea Basin. A Case Study from Howick, North-east England. pp. 189-202. Oxbow: Oxford.
178. Ward, I. & Larcombe, P. 2008. Determining the preservation rating of submerged archaeology in the\ post-glacial southern North Sea: a first-order geomorphological approach. Environmental Archaeology 13:59-83.
179. Ward, I., Larcombe, P. & Lillie, M. 2006. The dating of Doggerland — postglacial geochronology of the southern North Sea. Environmental Archaeology 11:207-218.
180. Ward, I., Larcombe, P., Firth, A. & Manders, M. 2014. Practical approaches to management of the marine prehistoric environment. Netherlands Journal of Geosciences 93:71-82.
181. Weerts, H., Otte, A., Smit, B. et al. 2012. Finding the needle in the haystack by using knowledge of Mesolithic human adaptation in a drowning delta. eTopoi: Journal for Ancient Studies 3:17-24.
182. Weerts, H. J. T., Borst, W. G., Smit, B. I., Smits, E., van der Plicht, J. & van Tongeren. O. F. R. 2015. Epilogue. Mesolithic human skull fragments of the Maasvlakte 2 artificial beach. In Moree, J. M. & Sier, M. M. (eds.) Interdisciplinary Archaeological Research Programme Maasvlakte 2, Rotterdam. BOORrapporten 566. pp. 399-413. Gemeente Rotterdam / Rijksdienst voor het Cultureel Erfgoed & Port of Rotterdam: Rotterdam.
183. Weninger, B., Schulting, R., Bradtmöller, M. et al. 2008. The catastrophic final flooding of Doggerland by the Storegga Slide tsunami. Documentia Praehistorica 35:1-24.
184. Wessex Archaeology 2009. Seabed Prehistory: Site Evaluation Techniques (Area 240) Existing Data Review. Archaeology Data Service: York.
185. White, M. J. & Schreve. D. C. 2000. Island Britain— Peninsular Britain: palaeogeography, colonisation and the Lower Palaeolithic settlement of the British Isles. Proceedings of the Prehistoric Society 66:1-28.
186. Zachos, J., Pagani, M., Sloan, L., Thomas, E. & Billups, K. 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292:686-693.
187. Zagwijn, W. H. 1979. Early and Middle Pleistocene coastlines in the southern North Sea Basin. In Oele, E., Schüttenhelm, R. T. E. & Wiggers, A. J. (eds.) The Quaternary History of the North Sea. Acta Universitatis Upsaliensis Symposia Universitatis Upsaliensis Annum Quingentesimum Celebrantis 2: 31-42.
188. Zagwijn, W. H. 1983. Sea-level changes in The Netherlands during the Eemian. Geologie en Mijnbouw 62:437-450.