Geological evolution of the Rhine-Meuse delta

(synopsis from Berendsen 2005)

The Rhine-Meuse system

Figure 1 The North Sea basin, and subdivision of the Dutch coast. The delta area (A) is characterized by large tidal inlets. The central coast of Holland (B) consists of barrier beaches with coastal dunes. The northern coast (C) consists of islands and tidal inlets to the 'Wadden Sea', where tidal flats occur. The southern North Sea Basin is shallow. On the sea floor elongated sand ridges occur that appear to be related to tidal currents (after Van de Meene 1994). Enlarge

The Holocene Rhine-Meuse deltaic plain is situated in the southeastern corner of the North Sea Basin (Figure 1).

The geological map of the Netherlands (Figure 2) shows, that approximately 45 % of the Netherlands consists of Holocene deposits, 50 % of Pleistocene deposits and in only 5 % of the area older rocks crop out.

Figure 2 Geological map of the Netherlands (after Berendsen 2004). Lithostratigraphy after De Mulder et al. (2003). The so-called perimarine area is characterized by fresh-water river sediments and extensive peat formation. River channels in this area are of the low-sinuosity meandering type or straight, and have a low width/thickness ratio of the sandbody. River banks consisting of clay and peat resist lateral erosion, and crevasse splays are abundant. Enlarge
Figure 3 Catchment area of the Rhine and Meuse. Enlarge

The present mean annual discharge of the Rhine is about 2200 m3/s; that of the Meuse (Dutch: Maas) approximately 250 m3/s (Table 1). Rhine discharge is divided among three distributaries: Waal (6/9 of total discharge), Nederrijn-Lek (2/9) and IJssel (1/9). The Rhine has a mixed snowmelt and rainfall discharge, with a peak discharge in the spring and a smaller peak during the early summer. The Meuse is a typical rainfed river. The Meuse (Maas) has been a tributary of the Rhine during most of geological history.

Table 1 Present discharge in m3/s of the rivers Rhine, Meuse, Rhone, Mississippi
Rhine Meuse Rhone Mississippi
Minimum 620 30 360 5600
Mean annual 2200 250 1670 12000
Maximum 13000 3000 13000 56000
Min/Max ratio 1:20 1:100 1:36 1:10

The combined catchment of the Rhine and Meuse is shown in Figure 3.

Tertiary and Pleistocene evolution

The North Sea Basin formed because of Mesozoic stretching related to the opening of the Atlantic Ocean. Subsidence increased considerably in the Quaternary. A Quaternary sequence of up to 1000 m thickness formed in the central part of the North Sea area (Figure 4).

Figure 4 Thickness of the Quaternary in the Netherlands. Enlarge
Figure 5 Pleistocene stratigraphy (after Berendsen 2004). Average July temperatures are essentially based on pollen analysis. Correlations with the marine oxygen isotope stages are provisional. Enlarge

The oldest sediments of the Rhine date from the Miocene when it was a small stream (Figure 6a) draining the Graben of the Lower Rhine Embayment. Marine deposition predominated in the Netherlands until the Early Pleistocene (2.5 million years ago).

Neogene uplift of the Rhenish Massif (Germany) and the Ardennes (Belgium) led to increasing drainage areas both for the Rhine and Meuse (Figure 6b, c). At the Pliocene/Pleistocene boundary the Rhine captured some main tributaries of the Saône-Rhône, e.g. the river Aare (Switzerland, Figure 6b, c), and extended its drainage area in the Alps. During the Pleistocene Holsteinian (marine isotope stage 11; Figure 5) the Oberrhein (until then a tributary of the Danube) was captured by the Rhine. The Saalian glaciation finally diverted the upper course of the Meuse via the Mosel to the Rhine.

Figure 6 Increase of the drainage area of the Rhine by stream capture. After Berendsen & Stouthamer (2001). Palaeogeographic situation during (a) the Miocene. The Rhine was a small stream, the Danube drained the Alps that were uplifted and folded. (b) Pliocene. The drainage area of the Rhine increased. The Alps drained to the Saône-Rhône. (c) Early Pleistocene. The Rhine extended its drainage area into the Alps. The Meuse extended its drainage into the Vosges mountains. (d) Present. The upstream part of the Meuse drainage was captured by the Mosel during the Saalian glaciation. Enlarge

During the Pleistocene practically the entire Netherlands became part of a subaerial delta formed by the rivers Rhine, Meuse, Scheldt, Elbe and Weser (Figure 7). During the Saalian (Figure 8a), approximately half of the Netherlands became covered by ice from the Scandinavian ice sheets. The Saalian ice cap significantly altered the landscape, forming 100 m high glaciotectonic ridges that are still important elements in the landscape, because they still partly control the width of the fluvial plain.

Figure 7 Palaeogeographic situation in the Netherlands during the Early and Middle Pleistocene. After Zagwijn (1974). The coastline changed from concave to convex in the Late-Tiglian (see Figure 6) as a result of a high supply of debris, even though the Roer Valley Graben and the Lower Rhine area were subsiding. After the Waalian supply by Baltic rivers ceased. Enlarge
Figure 8 Palaeogeography of the Rhine-Meuse delta during the Late Pleistocene (essentially after Zagwijn 1986). During the Saalian, half of The Netherlands became covered by land ice, and river flow became diverted. The ice formed 100 m high glaciotectonic ridges, that are still visible in the present landscape. During the Eemian interglacial, marine deposits were laid down in former glacial basins, and in the downstream parts of river valleys. During the Weichselian, most of the Netherlands became covered by eolian sands (so-called coversands). Enlarge

During the Early Weichselian, the Rhine and Meuse flowed westward through two valleys that are still visible in the morphology of the Late Weichselian surface (Figure 8d). The northern course through the IJssel valley became abandoned during the Weichselian Pleniglacial. By the end of the Weichselian glaciation (in which the ice sheets did not reach the Netherlands, Figure 8d) most of the Netherlands became covered with eolian sands (Figure 8d). At that time, both Rhine and Meuse flowed westward and discharged to the Atlantic Ocean via the Strait of Dover (in fact, this flow direction was already established during the Elsterian). Due to climatic changes in the Late-Glacial two terrace levels were formed, called the Lower Terrace and Terrace X. Both consist of braided-river deposits (Figure 9).

Figure 9 Climate changes and changes in fluvial style during the Late Weichselian and Holocene (after Berendsen, Hoek & Schorn 1995). Enlarge

A dry, windy climate enhanced eolian reworking of river sand and gave rise to extensive dune formation. These eolian dunes are generally referred to as 'river dunes' in the Dutch literature. These eolian dunes are up to 20 m high, and often are still visible in the present-day landscape.

Holocene evolution

The Holocene rivers follow an E-W course and cross the SE-NW trending tectonic structures of the Roer Valley Graben, Peel Horst en Venlo Graben (Figure 10). Differential subsidence at the Peel Boundary fault is approximately 2 m over the last 14.000 years (Cohen 2003). Gradient lines of channel belts have been deformed by post-depositional neotectonic movements (Stouthamer & Berendsen 2000).

Figure 10 Tectonic elements in the central Netherlands (Cohen 2003). The Peel Horst is relatively uplifted compared to the Roer Valley Graben and the Venlo Graben. The Peel Boundary Fault is the most active fault in the Netherlands. Enlarge
Figure 11 Sealevel rise and groundwater gradient lines (after Van Dijk et al 1991). Groundwater gradient lines are based on the dating of peat layers on the flanks of Younger Dryas eolian dunes, assuming that the peat approximately reflects groundwater levels. East of Gorkum, gradient lines are dominated by river gradients; west of Gorkum tides influenced the rivers. Holocene sealevel rise causing onlap, resulted in an upstream shift of the terrace intersection of Holocene deposits and the Kreftenheye Formation (Lower terrace). The shift of the terrace intersection approximately corresponds to the successive intersections of groundwater gradient lines with the Kreftenheye Formation. Enlarge
Figure 12 Shifting of the terrace intersection as a result of sealevel rise during the Holocene (Stouthamer & Berendsen 2000). The location of the terrace intersection of the Pleniglacial terrace (Kreftenheye Formation) and Holocene deposits shifted rapidly upstream between 8000 and 7000 yr BP. Then the upstream shift decreased as a result of the topographic high position of the Peel Horst. After 6000 yr BP relief had been leveled and the terrace intersection shifted relatively fast upstream again. Eventually the rate of upstream shift decreased as a result of decreasing sealevel rise. Enlarge

In the Early Holocene fluvial style changed from braided to meandering. At first, the meandering rivers were incised. In the western part of the present deltaic plain, aggradation started in the early Atlantic (after 8000 yr BP, Figure 11). The sea first invaded the mouths of the Pleistocene river valleys. Since the influence of sealevel rise was felt earlier in the lower western part of the country, clayey and peaty floodbasin deposits on top of the sandy Pleistocene subsurface are younger in an eastern direction (Figure 12).

The terrace intersection of the Pleniglacial terrace and Holocene deposits shifted landward as a result of Holocene sealevel rise. The rate of this onlap is shown in Figure 12, a diagram that is based on 14C-dated peat samples at the base of the Holocene. Sealevel rise slowed continuously, and this would have to result in an ever decreasing landward shift of the terrace intersection. However, it can be seen that the shifting of the terrace intersection first slowed approximately 6000 yr BP, and then increased again. This is a result of the local higher elevated Peel Horst.

In the near-coastal fluvial area, the fast rate of sealevel rise gave rise to a low-energy, narrow, anastomosing river pattern between approximately 7500 and 4000 14C yr BP (Figure 13). As sealevel continued to rise, this pattern extended further to the east, but it never crossed the Peel Boundary Fault. Individual channel belts consist of straight, ribbon-like sandbodies with a low width/thickness ratio (<15). As sealevel rise decreased, in the younger part of the Holocene, low-sinuosity meandering rivers reached further to the west.

Figure 13 Changes in fluvial style during the Holocene, and their relation to sealevel rise (after Törnqvist 1993, modified by Berendsen & Stouthamer 2001). Rapid sealevel rise, causing high aggradation rates, is believed to be responsible for the formation of a straight river pattern with large-scale crevassing in the west-central part of the delta, between approximately 7500 and 4000 yr BP. When the rate of sealevel rise decreased, meandering rivers migrated westward and large-scale crevassing stopped. Near the northern and southern margins of the delta, straight rivers did not develop, because here the sandy substrate consisting of coversand enhanced lateral migration, and meandering. Enlarge

In a longitudinal direction, spatial changes in fluvial style depend on gradient and substrate erodibility (Figure 14). In the area upstream of the terrace intersection of the Lower terrace and Holocene deposits, incising high sinuosity meandering rivers occur. Examples are found just across the German border. Downstream of the terrace intersection, the high-sinuosity pattern is maintained, although rivers are aggrading.

Figure 14 Spatial change in river morphology (after Berendsen & Stouthamer 2001). Meandering rivers occur near the terrace intersection, where easily erodible sandy deposits occur at shallow depth, and river gradient is still relatively high. Westwards, gradients decrease, as well as stream power, and river channels become straight. This longitudinal succession of river channel pattern shifted eastwards during the Holocene as a result of sealevel rise. Enlarge

In the eastern part of the delta, the sandy, easily erodible Pleistocene substrate is still at shallow depth, which enhances lateral (bank) erosion. Further downstream the river pattern changes into low-sinuosity meandering or straight (Figure 14). Here, the Holocene consists mostly of thick clay and peat layers, that resist lateral erosion. This leads to straight channels with a low width/depth ratio.

This longitudinal succession of river patterns existed also in the past, and shifted upstream as a result of Holocene sealevel rise. This means, that in the western part of the delta, where rivers are now straight, meandering channel belts may be found at greater depth. This relatively simple model is complicated by differences in substrate composition. In the marginal parts of the delta (the upstream area, and the northern and southern fringes) the substrate consisted of easily erodible Pleistocene fluvial sand or coversand.

Here, lateral erosion was common, and meandering fluvial systems developed, whereas in the central part of the delta, where the substrate consisted of clay, low-sinuosity meandering or straight rivers prevailed (Figure 13).

In general, channel belt width/thickness ratios decrease downstream (Figure 15). This is a result of decreasing width, as well as slightly increasing thickness (or depth of the channel).

Figure 15 Fence diagram depicting the longitudinal architecture of channel deposits and overbank deposits of the Schaik channel belt (number 150 on the map of Berendsen & Stouthamer 2001). Interpreted from lithological sections by Törnqvist et al. (1993). Enlarge
Figure 16 Longitudinal changes in channel belt width and channel width. West of coordinate x=115 (near Rotterdam) channel belt width equals channel width, indicating that lateral accretion is no longer important. The perimeter of the sandbody here is representative for the wet perimeter of the channel, which allows making discharge comparisons. Enlarge

West of coordinate x=115 channel belt width is virtually equal to channel width (Figure 16). This means that lateral accretion is this area is absent, and the channel belt sandbody perimeter is equal to the channel wet perimeter. Hence, channel dimensions can be used to compare discharge of individual channels.

Figure 16 also shows, that channel belt width of recent channel belts (Waal, Nederrijn-Lek) is larger than channel belt width of older channel belts. This is most likely a result of human influence: the number of coeval channels has decreased as a result of damming of older distributaries. This resulted in an increase of discharge of individual channels, and a greater channel belt width.

Channel belt thickness of the recent meandering distributaries (Waal, Nederrijn-Lek) is equal to channel depth. This suggests, that these meandering channel belts are a product of lateral accretion, and that vertical in-channel accretion is not important. Whether this also holds for the straight anastomosed rivers (that existed in the western part of the delta between 8000 yr BP and 4000 yr BP, Figure 14) is presently unknown. According to Makaske (1998, p. 55) the straight anastoming Columbia River in British Columbia (Canada) may show significant vertical in-channel accretion.

Avulsion

Avulsion (the formation of a new river channel, and abandoment of the old channel) was an important process in the evolution of the Holocene Rhine-Meuse delta (Stouthamer & Berendsen 2000). Approximately 90 avulsions occurred over the last 7000 14C yr, but only 30 of them can be considered as major avulsions. The location of avulsion sites (on a time scale of millennia) is influenced by the same factors that influenced the palaeogeographic evolution. The main factors are: sealevel rise, neotectonics and changes in discharge and sediment load. All avulsions presumably start with the formation of a crevasse splay. In many cases, large crevasse splays occur near avulsion sites. Especially between 4000 yr BP and 2000 yr BP, many avulsions seem to be related to differential tectonic movements of the Peel Horst and the Roer Valley Graben (Figure 17). The relatively fast subsidence in the area west of the Peel Boundary fault created new accommodation space, in an area where rivers still had a significant gradient. This enhanced the occurrence of nodal avulsions near the Peel Boundary Fault.

Figure 17 Sealevel rise, location of the avulsion sites in the Rhine-Meuse delta plotted on an E-W axis, and avulsion frequency during the Holocene (Stouthamer & Berendsen 2000). The spatial distribution of avulsion locations is mainly determined by: sealevel rise (7500-3700 yr BP), neotectonics (4900-1700 yr BP), increased discharge and/or within-channel sedimentation (2800-1000 yr BP), and human influence (after 1000 yr BP). Note that the tectonic elements cut this section at an angle. Most avulsions occurred in the central part of the delta and around the western fault zone of the Peel Horst (between coordinates 130 and 170). The avulsion frequency reached a maximum from 3000 to 1700 yr BP; lowest avulsion frequencies occurred from 5300 to 5000 yr BP, 3500 to 3000 yr BP, and after 1500 yr BP (from Stouthamer & Berendsen 2000). Enlarge
Figure 18 Period of existence of channel belts in the Rhine-Meuse delta (Berendsen & Stouthamer 2001). The diagram shows no significant trend in the length of the period of existence of the channel belts. The average period of existence is approximately 1000 14C yr, or 1100 cal yr. Enlarge

Avulsion frequency initially was determined by rapid sealevel rise (during the Atlantic). A second maximum of the avulsion frequency was reached between 3000 and 1700 BP, which may be related to increased discharge and/or within-channel sedimentation, or both. The period of existence of channel belts varied widely throughout the Holocene (Figure 18), but shows no significant trend over time and space. It remained on average constant at approximately 1000 14C yr. The average avulsion duration is ~325 yr. It seems to be constant over time, although it varied widely, from 'instantaneous' avulsion (< 200 yr) to gradual avulsion with a duration of over 1000 14C yr. This implies that the average interavulsion period can be estimated to be ~600-700 14C yr, or ~800-900 cal yr, but variation is considerable (Stouthamer & Berendsen 2001).

If avulsion duration and interavulsion period are constant, then the number of coeval channels in the Rhine-Meuse delta was essentially determined by avulsion frequency. A lack of data from other deltas in the world presently prohibits to determine whether this a general characteristic, or a phenomenon that only applies to the Rhine-Meuse delta.

Sealevel rise

Sealevel rise has had a significant influence on the development of the Holocene stratigraphy in the estuarine area. The rate of sealevel rise during the Early Holocene was very high: about 1 m/100 14C years, but gradually decreased during the Holocene. The rapid rise of sealevel before 6000 yr BP is caused by the melting of the Weichselian ice sheets; the slower rise of sealevel during the younger part of the Holocene is mainly due to isostatic subsidence. The Holocene transgression caused an overall stack in backstepping pattern (Figure 11), and peat formation moving inland. Groundwater level rise is also influenced by local differential tectonic movements of the Roer Valley Graben and the Peel Horst (Cohen 2003, Figure 19).

Figure 19 Groundwater level rise around the Peel Boundary Fault. Local groundwater curves show the effect of differential subsidence (Cohen 2003). Enlarge
Figure 20d Legend for Figure 20

The back-barrier intertidal and estuarine deposits were formed during two main periods of transgression: during the Atlantic (8000 - 5000 yr BP) and during the Subatlantic (3000 yr BP to the present). The deposits were previously known as 'Calais Deposits' and 'Duinkerke (=Dunkirk) Deposits' respectively. In recent years, this subdivison has been abandoned, because of difficulties in recognizing these units. Now, a lower clastic Member (the Wormer deposits) and an upper clastic Member (the Walcheren deposits) are recognised (de Mulder et al. 2003), based purely on lithostratigraphic criteria.

Figure 20 a
Figure 20 b
Figure 20 a, b, c Holocene palaeogeography of the Netherlands (after De Mulder 2003, and Berendsen 2004) at various moments during the Holocene. Approximately 5300 yr BP the oldest barrier beaches were formed, east of the present coastline. Approximately 3700 yr BP the coast was closed and extensive peat formation occurred in the back-barrier area. Approximately 1900 yr BP (Roman times) coastal erosion started in the north and in the southwest. At that time the Oude Rijn (=Old Rhine) was still the main distributary of the Rhine. Approximately 500-700 AD coastal erosion increased, and main drainage of the Rhine shifted to the southwest. Rhine and Meuse (=Maas) both emptied into the 'Meuse estuary' near Rotterdam. The Oude Rijn was finally dammed in 1122 AD. Enlarge 20a, 20b, 20c

Historical evolution

Humans have been present in the Rhine-Meuse delta for at least 200,000 years; however, human influence on the landscape was small until the Neolithic (6400 - 3650 yr BP), when forests were cleared and agriculture started on the natural levees, on the Younger Dryas eolian dunes and on higher Pleistocene sands.

Approximately 2000 yr BP, the northernmost distributary of the Rhine (Vecht, Figure 1) silted up and the Old Rhine (which had been the most important Rhine distributary since 5000 yr BP) became the northern border of the Roman Empire during most of Roman times (50 BC - 400 AD). Roman occupation left many archaeological traces in the substrate, and many villages were founded. The Romans even influenced locally the course of rivers by digging canals.

Human influence increased enormously during the Middle Ages, especially from 1100 AD onwards. A first step was the embankment of the rivers, which was completed around 1300 AD. During the same time period the Old Rhine (Kromme Rijn - Oude Rijn) was dammed near Wijk bij Duurstede in 1122 AD (for the location see Figure 1) and the river mouth of the Old Rhine degraded. The ebb-tide delta was eroded and the sand became available for transport to the coast. This stimulated the formation of the so-called Younger Dunes, that form dune ridges up to 40 m high. The damming of the Old Rhine distributary and some other channels, like the Hollandse IJssel (1285 AD) and the Linge (1307 AD), reduced the number of Rhine distributary to the present three, and there was a tendency for the main flow to shift southwestwards. The Waal gradually became more important, and both the Meuse (=Maas) and the Lower Rhine-Lek joined the Waal before debouching through a tidal inlet just south of the present Nieuwe Maas-Rotterdam Waterway.

Figure 21 Water levels and tidal amplitude on the Lower Rhine-Lek (Berendsen 1982). The gradient lines represent highest high water, normal river level and lowest low water, measured over the period 1951-1960. The GTS line (Gradient line of the Top of the Sand) corresponds approximately to bankfull discharge water levels. Tidal influence at high discharge reaches 70 km upstream, to Jaarsveld. At low discharge tides reach 90 km inland. The upper histograms show tidal amplitude at different locations, with varying discharge. The influence of tides reaches further inland at low discharge than at high discharge. The lower histograms show that tidal amplitude on the river has increased over the last century. This is a result of the widening and deepening of the Rotterdam Waterway. Enlarge

The digging of the Pannendens Canal (PK in Figure 1) in 1707 AD significantly altered the discharge distribution over the Rhine distributaries. From then on the discharge distribution has remained the same. The Haringvliet (Figure 1) soon became the main outlet. In recent years this tidal inlet has been closed off by a dam, forcing the main flow through the Rotterdam Waterway (Figure 1), which was dug in 1872 AD.

The embanking of the rivers opened the possibility of draining and reclaiming the peat bogs in the western part of the deltaic plain. In the Late Middle Ages this peat was excavated and used for fuel or to extract salt for the population of the growing cities. Since groundwater levels were high, these peat excavations rapidly turned into lakes. Storms enlarged the lakes by wave action. In the long run, this created a dangerous situation, and many lakes were pumped dry successively, especially between 1600 and 1900 AD. One of the best known examples is the Haarlemmermeerpolder (the location of Amsterdam airport), the first lake that was pumped dry in 1852 AD using three steam engines.

The construction of groynes, especially since 1850 AD, has caused narrowing and deepening of the river beds, thus enhancing the shipping industry. When the entrance to Rotterdam harbour was widened, river water levels were lowered, and tidal influence is now experienced about 20 km further upstream (Figure 21) compared to a century ago.

References

  1. Berendsen, H.J.A. (1993), The Weichselian Aftermath in the Netherlands. Day 2: Fluvial Deposits. In: Mulock Houwer, J., W.F.H. Pilaar & Th. Van De Graaff-Trouwhorst Eds. (1993), The Netherlands: Sandbody geometry and sequence stratigraphy of Fluvial and Coastal Deposits in the Holocene Rhine/Meuse Delta System. Fieldtrip, AAPG International Conference and Exhibition, The Hague, Oct. 20-23, 1993, p. 17-22 and 33-60.
  2. Berendsen, H.J.A. (1993), Sandbody geometry and stratigraphy of fluvial deposits in the Holocene Rhine/Meuse delta. Berichten 19, 64 p.
  3. Berendsen, H.J.A., W.Z. Hoek & E.A. Schorn (1995), Late Weichselian and Holocene river channel changes of the rivers Rhine and Meuse in the central Netherlands (Land van Maas en Waal). Paläoklimaforschung / Palaeoclimate research Bd. 14, Special Issue ESF Project European Palaeoclimate and Man 9, Ed. B. Frenzel. p. 151-171.
  4. Berendsen, H.J.A. (1998), Birds-eye view of the Rhine-Meuse delta. World Deltas Symposium, New Orleans. Journal of Coastal Research 14 (3), p. 740-752.
  5. Berendsen, H.J.A. (2004), De vorming van het land. Inleiding in de geologie en de geomorfologie. Fysische geografie van Nederland. Vierde geheel herziene druk met CD-ROM. Assen: Koninklijke Van Gorcum.
  6. Berendsen, H.J.A. & E. Stouthamer (2000), Late Weichselian and Holocene palaeogeography of the Rhine-Meuse delta. Palaeogeography, Palaeoclimatology and Palaeoecology 161 (3/4), p. 311-335.
  7. Berendsen, H.J.A., E.L.J.H. Faessen, A.W. Hesselink & H. Kempen (2001), Zand in banen - Zanddiepte-kaarten van het Gelders Riverengebied met inbegrip van de uiterwaarden. Sand-depth maps of the eastern part of the Rhine-Meuse delta (with a summary in English). Arnhem: Provincie Gelderland, in samenwerking met Rijkswaterstaat, Waterbedrijf Gelderland en Universiteit Utrecht, 53 p. Coloured maps.
  8. Berendsen, H.J.A. & E. Stouthamer (2001), Palaeogeographic development of the Rhine-Meuse delta, The Netherlands, Assen: Koninklijke Van Gorcum, 268 p. 3 coloured maps, CD-ROM.
  9. Berendsen, H.J.A., & E. Stouthamer (2002), Palaeogeographic evolution and avulsion history of the Holocene Rhine-Meuse delta, the Netherlands. Netherlands Journal of Geosciences / Geologie en Mijnbouw 81 (1), p. 97-112.
  10. Berendsen, H.J.A. (2004), De vorming van het land - Inleiding in de geologie en de geomorfologie. Fysische geografie van Nederland. Assen: Koninklijke Van Gorcum, 410 p.
  11. Berendsen, H.J.A. (2005), The Rhine-Meuse delta at a glance. 8th International Conference on Fluvial Sedimentology, Delft, the Netherlands. Mid-conference excursion guide, August 10, 2005. Download as PDF.
  12. Cohen, K.M. (2003), Differential subsidence within a coastal prism. Ph.D. Thesis, Universiteit Utrecht. KNAG/Faculteit Ruimtelijke Wetenschappen, Netherlands Geographical Studies, 172p.
  13. De Mulder, E.F.J., M.C. Geluk, I. Ritsema, W.E. Westerhoff & Th. E. Wong (2003), De ondergrond van Nederland. Geologie van Nederland, deel 7. Utrecht: Nederlands Instituut voor Toegepaste Geowetenschappen TNO.
  14. Maarleveld, G.C. (1955), Fluvioglaciale afzettingen in Midden Nederland. Tijdschrift van het Koninklijk Nederlands Aardrijkskundig Genootschap, Tweede Reeks 72, p. 48 58.
  15. Maarleveld, G.C. (1956), Grindhoudende Midden-Pleistocene sedimenten. Dissertatie, Utrecht. Ook verschenen als: Mededelingen van de Geologische Stichting, Serie C-VI, 6.
  16. Maarleveld, G.C. (1981),The Sequence of ice-pushing in the Netherlands. Mededelingen van de Rijks Geologische Dienst 34, p. 2-6.
  17. Makaske, B. (1998), Anastoming rivers - Forms, processes and sediments. Ph.D. Thesis, Universiteit Utrecht. KNAG/Faculteit Ruimtelijke Wetenschappen, Netherlands Geographical Studies 249, 287 p.
  18. Middelkoop, H. 1997. Embanked floodplains in the Netherlands. Geomorphological evolution over various time scales. Ph.D. Thesis. KNAG / Faculty of Geogr. Sciences, Utrecht University: Nederlandse Geografische Studies 224, 341p.
  19. Stouthamer, E. & H.J.A. Berendsen (2000a), Factors controlling the Holocene avulsion history of the Rhine-Meuse delta (The Netherlands). Journal of Sedimentary Research 70 (5), p. 1051-1064.
  20. Stouthamer, E. & H.J.A. Berendsen (2000b), Avulsion frequency, avulsion duration and interavulsion period of Holocene channel belts in the Rhine-Meuse delta, The Netherlands. Journal of Sedimentary Research, 71 (4), p. 589-598.
  21. Stouthamer (2001), Holocene avulsions in the Rhine-Meuse delta, The Netherlands. Ph. D. Thesis, Utrecht University. Netherlands Geographical Studies 283, 224 p.
  22. Törnqvist, T.E. (1993a), Fluvial sedimentary geology and chronology of the Holocene Rhine-Meuse delta, The Netherlands. Ph. D. Thesis, Utrecht University, 169 p.
  23. Törnqvist, T.E. (1993b), Holocene alternation of meandering and anastomosing fluvial systems in the Rhine-Meuse delta (central Netherlands) controlled by sealevel rise and subsoil erodibility. Journal of Sedimentary Petrology, 63, p. 683-693.
  24. Törnqvist, T.E. (1994), Middle and late Holocene avulsion history of the River Rhine (Rhine Meuse delta, Netherlands). Geology 22, p. 711 714.
  25. Törnqvist, T.E. (1998), Longitudinal profile evolution of the Rhine-Meuse system during the last deglaciation: interplay of climate change and glacio-eustasy? Terra Nova 10 (1), p. 11-15.
  26. Törnqvist, T.E., M.H.M. Van Ree & E.L.J.H. Faessen (1993), Longitudinal facies architectural changes of a Middle Holocene anastomosing distributary system (Rhine Meuse delta, central Netherlands). In: C.R. Fielding, Ed. Current research in fluvial sedimentology, Sedimentary Geology 85, p. 203 219.
  27. Van De Meene, E.A., J. Van Der Staay And Teoh Lay Hock (1979), The Van der Staay suction-corer - a simple apparatus for drilling in sand below groundwater table. Rijks Geologische Dienst, Haarlem: 1-15.
  28. Van der Woude, J.D. (1983), Holocene palaeoenvironmental evolution of a perimarine fluviatile area. Geology and palaeobotany of the area surrounding the archeological excavation at the Hazendonk river dune (Western Netherlands). Analecta Praehistorica Leidensia 16, p. 1 124.
  29. Van Dijk, G.J., H.J.A. Berendsen & W. Roeleveld (1991), Holocene water level development in The Netherlands' river area; implications for sealevel reconstruction. Geologie en Mijnbouw 70, p. 311-326.
  30. Weerts, H.J.T. & H.J.A. Berendsen (1995), Late Weichselian and Holocene fluvial palaeogeography of the southern Rhine-Meuse delta (The Netherlands). Geologie en Mijnbouw 74, p. 199-212.
  31. Zagwijn, W.H. (1974), The Paleogeographic Evolution of The Netherlands during the Quaternary. Geologie en Mijnbouw 5, p. 369 385.