The Evolution of the Coastal Dunes in the Western Belgian Coastal Piain

[Die Entwicklung der Küstendünen in der westlichen Küstenebene von Belgien] K u r z f a s s u n g : Zwei Hauptdünengebiete kommen im Gebiet westlich von Nieuw­ poort vor: die Jüngeten Dünen entlang der heutigen Küstenlinie sowie die Älteren Dünen, die in det Nähe der französischen Grenze isoliert in der Küstenebene auftreten und ebenfalls küsten­ parallel verlaufen. Beide Systeme wurden geologisch und paläontologisch untersucht. Es wird vermutet, daß das Ältere Dünensystem in der Zeit vor 4300 v. h. entstanden ist. Ein Streifen weiterer älterer Dünensedimente (det jedoch nicht so alt wie das Ältete Dünensystem ist) wurde, ebenfalls nahe der französischen Grenze, untet den Jüngeten Dünen gefunden. Das Alter dieser Sedimente liegt zwischen 2800 und 900 v. h. Bei det Entstehung det Jüngeten Dünen wurden zwei Subphasen unteischieden. Vor dem 14. Jahrhundert bildete das Dünengebiet eine relativ ebene Fläche. Ihre Entstehung markiert das Ende einer aktiven äolischen Phase, die im 11. Jahr­ hundert n. Chr. wahrscheinlich als Folge damals verstärkt auftretender Sturmfluten einsetzte *) Address of the authot: Drs. R. DE CEUNYNCK, Geologisca Instituut, R. U. Gent, Krijgslaan 281, B — 9000 Gent. 34 Rudolf de Ceunynck (Dunkerque-3 Transgression). Mit dem Ende des 14. Jahrhunderts begannen sich gtoße Sichel­ dünen zu bilden, die Teile det früheren Landoberfläche begruben. Einige von ihnen bewegen sich noch heute. Die Entwicklung des Jüngeren Dünengebietes bei De Panne wird eingehend beschrieben, einschließlich der Vegetationsgeschichte des Dünengebietes und der anschließenden Küsten­ ebene. Einige Aufmerksamkeit wird auch der Lage früherer Küstenlinien gewidmet.

Here the evolution of the Younget Duneatea neat De Panne is described in more detail including aspects of the vegetational history of the dune area and adjacent coastal plain.Some attention is also diawn on the position of fotmet coastlines.

Introduction
In the last few years geological and pollenanalytical investigations have been carried out in the dunes of the western Belgian coastal plain, mainly in the area west of Nieuwpoort (LEBBE & DE CEUNYNCK 1980, DE CEUNYNCK & THOEN 1981).Two distinct dune systems exist in that area: a) The "Older Dunes" of Adinkerke, a dune ridge with less pronounced topo graphy running parallel to the coast.The present day location of the Older Dunes suggests that at one time the coast was situated more to the south of the actual one.Two radiocarbon dates of the base of an important peat layer situated at the landwards side of the Older Dunes yielded 4270 ± 65 B.P. (IRPA 590) and 4300 ± 65 B.P. (IRPA 589) 1 .The fact that the peat layer associated with it is absent north of these dunes where instead beach deposits are found indicates that the coastline was at that time situated just north of the Older Dunes, 2,8 km inland of the present coastline (fig.1).It also follows that the Older Dunes are older than 4300 B. P.
b) The "Younger Dunes" along the present coast, generally showing high parabolic dune forms and usually considered not to be older than the 9th century A.D. (DEPUYDT 1967).
The following summarizes the results of geological and palynological investigations carried out in the area covered with the Younger Dunes, mainly west and south of De Panne.1).To avoid confusion with the Older Dunes of Adinkerke these sediments will be designated informally as "older dunes of De Panne".Radiocarbon dating of a peaty layer within these sediments yielded 1965 ± 110 B. P. (Hv 9136) for the top and 2660 ± 100 B. P. (Hv 9137) for the base (fig.2).Earthenware of the Iron Age is associated with this peaty layer.Until now, no older peaty layers or humic former land surfaces have been found in the area.However dune formation probably started earlier, since the earliest date -2660 B. P.
-reflects only the beginning of a stabilisation phase of the dunes.A reconstruction of the Iron Age dune surface, partly hypothetical, is represented in fig. 2. It is based on 1) All tadiocarbon dates ate listed in table 1 with their calibated ages and some of them also with an estimated age based on additional atchaeological and/of histotical evidence.stiatigiaphical and archaeological evidence.Pollen analysis of the peaty layer revealed that most of the time the area was covered with shrub vegetation consisting mainly of Juniperus.No Hippophae -the dominant shrub of present-day dune vegetationwas found (DE CEUNYNCK & THOEN 1981).The relative importance of Quercus (13 to 22 %) points to woody patches in the vicinity.Together with the stratigraphical evidence this suggests that the coastline at the time was situated at least 1 km north of the actual one (fig.2; 2300 B. P.: climax of shrub and wood vegetation in the dune area).The Romans settled on the same dune area and Roman earthenware occurs very frequently on top or within the "older dune sediments of de Panne".Most of the previous mentionned shrub and wood vegetation was destroyed as a result of renewed eolian activity during The Roman period (DE CEUNYNCK & THOEN 1981).However this activity did not result in the deposition of large amounts of dune sands.Only within the northetn part of the dune sediments a layer of sterile sands clearly seperates the Iron Age and Roman archaeological layers.Repeated analyses of the buried Roman dune surface did unfortunately not yield any pollen.
Just before the beginning of our era the sandy tidal flat era south of the "older dunes of De Panne" progressively changed in a salt marsh.Then a reed peat comparable with the northern German "Darg" (eft.OVERBECK 1975) developed on most of the salt marshes just south of the dunes.This demonstrates a declining marine influence in that particular area.The base of the reed peat was dated 2080 ± 55 B. P. (IRPA 582; DO 2 fig. 1 & 3).About 270 A. D. the Romans left the area because of the pressure by German tribes and above all because of the increasing marine influence in the area (THOEN 1978).Most of the presentday Belgian coastal plain was then flooded by the sea; this event is traditionally called the Dunkerque 2-transgression (TAVERNIER et al. 1970).Consequently the area south of the "older dunes of De Panne" changed again in a tidal flat-salt marsh area.The Dunkerque 2-transgression destroyed most of the "older dunes" except the ones near De Panne and the "Older Dunes" of Adinkerke.The pollendiagram DO 2 (fig.3) clearly demonstrates the events described above.The clayey part (-5,47 to -5,66 m below surface) represents sediments of the Dunketque 2transgression and is characterized by a high content of tree pollen (max.45 %) and in a lesser degree by Chenopodiaceae (15-19%).Dinoflagellates, foraminifers and reworked Tertiary and Cretaceous pollen are frequent but not indicated in the diagram.Within the basal peat layer mainly pollen of Gramineae and Cyperaceae are found.The presence of Chenopodiaceae suggests that pan of the area still was a salt marsh.The small quantities of Juniperus and Hippophae and possibly Myrica and Calluna could have originated from the dune area.During the 5th century A. D. marine influence decreased and a reed peat started to grow again on parts of the salt marsh area (radio carbon date 1630 ± 55 B. P.; IRPA 581; also 1470 ± 50 B. P., IRPA 580).At about the same time a Merovingian commercial post was probably established in the "older dunes of De Panne" as is indicated by the discovery of Merovingian coins and fibulae.Karolingian fibulae have been found as well (TERMOTE 1984).We believe that this period marks the onset of a stabilisation phase of the remnants of the older dunes of De Panne lasting at least until the 11th century.The dune surface of that period is Tab.1: Radiocarbon dates with calibrated age according to KLEIN et al. (1982) and fot the samples younger than 2100 B. P. also with an estimated age based on additional essentially the same as the Roman surface, but local eolian disturbances are not excluded.Pollenanalytical evidence indicates that in part of the dune area an Ericaceaeheathland had developed.
The growth of the upper peat layer in the former salt marsh area (fig.3) stopped at about the same time when man started to use the polders for grazing sheep and cattle.The polders were then no more than high salt marshes with restricted peat areas near the dunes.This "first" landuse of the polders has been historically and archeologically dated as of the 9th and 10th centuries A. D. (VERHULST 1967).At DO 2 a small bone of a cow (!) was found on top of the peat surface which in tutn was covered with dune sediments of the Younger Dunes.

The younger dunes west and south of de Panne (1000 A. D. onwards)
The so-called Dunkerque 3-transgression (11th century A. D.;TAVERNIER et al. 1970) most probably caused further etosion of the dunes of that time.But on this occasion latge amounts of sands were blown inland in the shape of moving dunes.This process means that sites situated inland could have been buried underneath a moving dune for instance 100 years or even more after the original formation of that particular dune, depending on the velocity of movement and the distance of the site from the point of origin of that dune (the beach of that time).So it is not surprising that part of the polders became progressively buried underneath a few meters of dune sands most probably during the 12th and 13th century A. D. i. e. almost two centuries after the Dunkerque 3-transgression seeing that the moving dunes first had to ctoss the dune   1).Both dates suggest that the sites were buried underneath dune sands at about the same time during the transi tion of the 14th to the 15th century A. D. Since these sites are situated at some distance of the sea, it is inferable that the moving dunes that caused the burying of these sites wete formed earlier.We tentatively link these events with a new important eolian phase related with the present day parabolic dunes.During this phase more of the poldets became coveted with dune sands (fig.2).Furthermore the sea eroded the dune area west of De Panne until the present northern border of the "older dune sediments" was reached.
Historical, archeological and radiometric evidence suggest that since then the parabolic dunes never totally ceased moving.Even today some of the parabolic dunes are still progressing at a rate of a few meters a year.Furthermore the natural vegetation was disturbed by the intensive grazing with sheep and cattle and by the burrowing activities of rabbits and many blow-outs were formed obliterating the original topo graphy.To complicate things even more, the direction of displacement of the parabolic dunes (N 72° to N 75°) differs only slightly from the general direction of the coastline (N 64°).Togethei with the distuibances induced by man this creates a complex internal topography of the Younger Dune area with parabolic dunes, large almost uncovered transversal dune ridges and other secondary dune forms.This makes it difficult to identify seperate series of parabolic dunes as a result of which one would eventually be able to identify different phases of eolian activity and dune stability.However it is known that by the end of the 18th century man tried to stabilise the areas thieatened by moving dunes by among other things the planting of trees.He only succeeded partially.
The coastline had reached its present location already at the end of the 16th century A. D.; this can be deduced from maps of that time.However thete are indica tions of periodical small scale changes of the coastline within the last centuries (sea also DE MOOR 1979).Only 10 years ago for instance the coast between Koksijde and Oostduinkerke was still prograding, but at present this area is again eroding.

2.
The Period between 800 B.C. and 1000 A.D. Already in 1951, MOORMANN indicated the presence of older dune sediments beneath the sediments of the Younger Dunes of De Panne.The extension of these sedi ments was studied by LEBBE & DE CEUNYNCK (1981) (fig.

Fig. 3 :
Fig. 3: Pollendiagram DO 2 of the salt maish sediments and peat layers just south of the oldet dunes of De Panne and now coveted with the Younger Dunes (localisation on fig.1); absolute heigth of the surface +8.5 m O. P., pollensum does not include spores and aquatic pollen.

cross section fig.2 ///////// older dune sediments beneath younger dunes
DEPUYDT, 1967 archeological evidence; DEPUYDT =DEPUYDT, 1967 area (fig.2).Afterwards, the dunelandscape consisted of an almost level plain with an absolute heigth of about 6 to 7 m O. P. 2 (area west and south of De Panne) most likely bound by a dune ridge to the sea.A reconstructed coastline of the 13th century is indi cated on fig.1.It is based on archeological, historical and geological evidence(modified  after TERMOTE 1984).The previously mentionned level plain has been encountered on different occasions in corings or in exposures and is marked by a buried soil.This former dune surface is indicated on fig.2(14th century surface).Pollenanalyses of this former landsurface reveals that at first vegetation was predominantly herbaceous.Very rapidly a shrub developed with Hippophae rhamnoides as predominant species in the drier areas and Salix arenaria as the dominant species in the humid areas.The surface has been dated on two occasions: 550 ± 75 B. P. (DEPUYDT 1967a; localisation on fig.2) and 590 ± 50 B. P. (IRPA 579, localisation = DO 1 on fig.