HYDROGEOLOGICAL AND HYDROGEOCHEMICAL INVESTIGATION OF THE DUNE AREA AND ADJACENT LOW POLDERS AT WENDUINE-UITKERKE, FLEMISH COASTAL PLAIN

K. WALRAEVENS, S. CLAYS, A. HERMANS & M. VAN CAMP

Ghent University (), Laboratory for Applied Geology and Hydrogeology, Krijgslaan 281 - S8, B - 9000 Gent, Belgium, [email protected]

ABSTRACT

The hydrogeology and hydrogeochemistry of the dune area at Wenduine and the polders behind it have been investigated. Of special interest are the lowlying polders at Uitkerke, that are situated at the flow lines’ end. For the purpose of a nature restoration project, that will be implemented soon, a hydrogeological investigation is currently being carried out in the core of these lowlying polders. Insight in the overal hydrogeological functioning of the system has been obtained beforehand. The general hydrogeological setting will be the subject of this contribution.

The geometry of the groundwater reservoir in the dune area and polders has been established. Groundwater flow has been determined. The groundwater chemistry has been studied by means of resistivity loggings in boreholes and by groundwater analysis. The interpretation of the both vertically and laterally strongly differentiated groundwater salinity is based on the hydrogeological history of the area.

STUDY AREA: WENDUINE- The dune belt of the North Sea UITKERKE between Wenduine and has a rather restricted width, no more Location and morphology than 50 m. Between Wenduine and De Haan the dune belt is far more The study area is situated in the developed and reaches a width of 1 eastern Flemish Coastal Plain km. (Belgium) and is to be found on map Further on, there is a connection with 4/7-8 (1:25.000) of the NGI (National the interior dune belts of Klemskerke Geographic Institute of Belgium) (fig. 1 ). and Vlissegem in the south. These The study area lies within the dunes have a varied relief of pans and municipalities of Wenduine (De Haan), parabole dunes, which are mostly Nieuwmunster () and fixated by plantations. Uitkerke (Blankenberge) and contains the extensive agricultural Polder of The polders within the study area Uitkerke. are of two kinds. First, those which are located on the top of former creeks (the so called “creek ridges”): the Fig. 1 - Location of the study area with indication of profiles A-A’-A” and B-B’-B”

Dunkirk-clay was deposited on the rich sands of the Member of Vlierzele sand of the creek. Second, the lower (Formation of Ghent) form the Tertiary clay-on-peat sediments (the so called surface in the rest of the study area. “pool grounds”). Both types form part The transition from the Formation of of the “’ Old Land”. Ghent to the underlying Formation of is to be found between De Haan Geological setting and Wenduine.

Based on well logging and Quaternary borehole descriptions of DE BREUCK During the Eemian interglacial age, (1966, 1969), DEVOS (1984), VAN the deposit of Oostende took place. Its HAECKE (1998), CLAYS (1999) and top and base are characterized by silex HERMANS (1999), the nature and and gravel parts, the deposit itself is level of the Tertiary surface and the mainly sand with shell fragments. Late structure and thickness of the Pleistocene sediments are, Quaternary deposits in the area of De consecutively, the deposit of Uitkerke Haan-Blankenberge were established, (sand) and the deposit of Wenduine which gives us insight in the geological (sand and gravel) (DE BREUCK, DE setting. MOOR & MARECHAL, 1969).The Pleistocene age was followed by a Tertiary period of peat-growth. (BAETEMAN, The Tertiary surface of the Polder 1987).The latter deposit was eroded of Uitkerke has the form of a bowl, during Atlanticum age (Flandrian originated by Early Pleistocene erosion Transgression; AMERYCKX, 1961) in (Saalian). Tertiary sediments were the centre of the Polder of Uitkerke. eroded in the centre of the bowl just The deposit of that transgression, the until the clay of the Member of deposit of Calais, can be divided in the Merelbeke (Formation of Ghent), in its eastern coastal area in two deposits: surroundings the clay-loam of the the sandy deposit of Houtave (base) Member of (Formation of and the deposit of Zuienkerke (top), Ghent) was preserved. The glauconite- which has a sandy character, but also contains some clay (DE BREUCK, DE the semi-pervious layer A’ is thicker MOOR & MARECHAL, 1969). At the due to the existence of the end of the Atlantic age, the peat of Nieuwmunster peat (A”), that reaches Nieuwmunster was formed behind the a thickness in polderland of some 2-3 dune belt (DEVOS, 1984) which was m. Pervious layer B contains the creek removed during later transgressions, sands of Dunkirk, together with known as the Dunkirk Transgressions. reworked parts of the Nieuwmunster Of most influence was the Dunkirk II peat, eroded by the creeks. Thickness transgression which formed creeks at varies up to 6 m. Semi-pervious layer Wenduine and Blankenberge, where B’ is composed by the Dunkirk clay, the peat of Nieuwmunster was and also forms the top layer in the removed. Nowadays, on these sites polderland, where it forms one layer remain ‘creek ridges’ (clay on sand), together with A” . Pervious layer C is and sites formerly higher situated, are the sand of the recent dune belt, where at a lower level (subsidence through it also forms the phreatic aquifer. The clay on peat). This phenomenon called level of the groundwater table is relief inversion was of a great subjected to seasonal fluctuations due importance in the study area. More to the variations in infiltrated rain water. recent relief changes such as peat- In the higher dunes which can reach a digging during Medieval Ages up to the level of + 20 m TAW, the water table late 19th century and clay-digging for is situated between 5 and 10 m of bricks up to the middle 20th century depth. In the lower and plain also took place. polderland the groundwater table is to be situated in B’ at approximately 0.75 Characterization of the groundwater m of depth (+ 2.5 m TAW). reservoir

The Quaternary sediments in the METHODOLOGY study area form a layered groundwater reservoir with a thickness of Observation wells approximately 30-35 m (fig. 2). Three aquifers can be determined (A, B, C), The hydrogeological survey carried separated by 2 semi-pervious layers out in 1998-1999 was based on the (AVA” and B’). In the polder area, B installation of observation wells. cuneates into disappearance. Between Washed borings throughout the study De Haan and Wenduine, and in area took place, in order to put up a Uitkerke (Blankenberge) the base of network of observation wells. In the pervious layer also contains the October 1998, this network contained Tertiary sand of the Member of 14 sites where pairs or triplets of Vlierzele. observation wells had been installed. The screens of these wells are situated Pervious layer A is mainly at different depths, in order to establish Pleistocene sand, covered with the vertical groundwater flow patterns. Holocene sands of Calais. Its thickness Screens were put at the base of the varies between 10 and 25 m. Some groundwater reservoir (25 to 34 m of local clayey intercalation in this layer depth), at the top of the ground water cannot be seen as proper impervious reservoir (5 to 10 m of depth) and in layers. Semi-pervious layer A’ contains the polderland at the groundwater the clayey/loamy top of the deposit of table (less than 1m of depth). At the Zuienkerke (deposit of Calais). Locally, piezometer tube, also miniscreens were installed to enable sampling at taken from all screens and miniscreens various depths. along the profiles throughout the study area, in order to interpolate the results. Resistivity logging

Long Normal (LN) and Short RESULTS Normal (SN) resistivity was measured in the open borehole with the Horizontal and vertical groundwater equipment constructed at the flow Laboratory for Applied Geology and Hydrogeology, in normal configuration. Through monitoring the hydraulic A resistivity log gives in fresh water head in the observation wells in the conditions a detailed view of the study area, horizontal and vertical composition and the thickness of the groundwater flow was established for logged layers. However, in saline both an autumn and early spring circumstances, these lithological situation. The dune area at De Haan- properties are overruled by the effects Wenduine is the recharge area and of water quality variations. In these discharge takes place in the lowlying conditions, the information obtained polderland. This trend is most obvious form LN-resistivity can serve in at shallower depths, but also persists establishing a groundwater salinity log. at the base of aquifer A. A watershed The salinity of the study area was could be determined which is studied along profiles, in which the influenced by seasonal changes. Tidal information of the LN-resistivity log was influence on the groundwater level has used. The LN-resistivity is a good been observed inland until a distance approximation of the true formation of more than 700 m of the high water resistivity (pt).The relation between the line. formation resistivity (pt) and the resistivity of the pore water (pw) for the Salinization saturated zone of the groundwater Salinization of the study area took reservoir is given by: place during the Flandrian and Dunkirk marine transgresions. Due to the Pt F . pw infiltration of rain water, freshening of (in which F = formation factor) the ground water reservoir occurred, and is most advanced in the recharge The formation factor is not a constant, area of the dunes. In the polderland, but in comparison with the water where infiltration is inhibited by the resistivity it differs little. Assuming a clay-peat-loam sediments, seawater constant formation factor=4, each conditions have been largely variation of pt will be interpreted as a preserved. Both profiles (fig. 3) variation in pw. illustrate this situation: the dune area has a moderately fresh water body Groundwater analyses underneath, which saliens inward polderland. In profile A-A’-A” , this Pumping on filters and miniscreens fresh/salt water distribution is disturbed (using a vacuum pump) until pH and at SB9, at the polder-dunes contact: a conductivity remain constant, allows to tongue of fresher water cones up obtain good ground water samples of towards polderland and another the study area. Sampling included tongue with brackish water cones into filtering and fixation. Samples were the dunes. This phenomenon is induced by both the lithological reservoir that resulted from the marine constitution and groundwater flow. transgressions. Profile B-B’-B” shows clearly the CONCLUSIONS existence of a moderately to weakly fresh water body underneath the The hydrogeological history of the dunes, which runs along two paths study area began with the toward the polderland: via both transgressions (Flandrian, Dunkirk) aquifers A and B. that salinized the ground water reservoir due to the infiltration of Distribution of groundwater types seawater. At present time, this salinization has already been reduced Using Stuyfzand’s classification to a large extent underneath the (STUYFZAND, 1986) the groundwater dunes, but still persists in polderland. type of each analysis was determined. Flowlines run from the recharge area Fig. 4 show the results for the two of the dunes in two directions: towards profiles. These profiles are comparable the sea (rather restricted) and towards to the resistivity profiles. the polderland, which is the main discharge area. The groundwater Fresh rainwater infiltration is quality distribution confirms this diluting the salt porewater. In the situation. dunes, this dilution is obvious. On the contrary, in the lowlying polder, almost the whole groundwater reservoir is stil REFERENCES containing salt water of the NaCI watertype. The cation exchange code AMERYCKX, J. (1961). La genese des ‘+’ means that cation exchange as a poldres maritimes belges. De result of freshening took place. Cation Aardrijkskunde, 13, pp. 5-18. exchange is stronger in the shallower BAETEMAN, C. (1987). Ontstaan en layers of the groundwater reservoir. evolutie van de kustvlakte. In: The Na+ of the clay is exchanged with Thoen H. De Romeinen langs de the Ca2+ of the water, which renders a Vlaamse kust, Gemeentekrediet NaHCOs watertype (WALRAEVENS, van België, Brussel, pp. 181. 1990). In the recharge area, where CLAYS, S. (1999). Hydrogeologisch en freshening has progressed further, the hydrogeochemisch onderzoek CaHC03 type of the infiltrating water is van het westelijk deel van de preserved. This freshening process Uitkerkse polder bij Wenduine. can be seen very well in both profiles M.Sc. Thesis, RUG, Ghent, pp. along the flowlines. While in the dunes 125 the process has already been DE BREUCK, W., DE MOOR, G. & terminated, the freshening in the polder MARECHAL, R. (1969). only started at its edges. In profile A- Lithostratigrafie van de Kwartaire A’-A” there is a clear upconing towards sedimenten in het Oostelijk the polders of water in the first phase Kustgebied. Natuurwet. Tijdschr., of freshening, in profile B-B’-B” the two 51, pp. 125-137 tongues of fresher water are also DE MOOR G. & DE BREUCK. W. present. In the core of the lowlying (1969). De freatische waters in polders, a negative cation exchange het Oostelijk Kustgebied en in de code is still pointing towards the Vlaamse Vallei. Natuurwet. salinization of the groundwater Tijdschr., 51, pp. 3-68 + ann. DEVOS, J. (1984). Hydrogeologie van het duingebied ten oosten van De Haan. Ph.D. Thesis, RUG, Gent, pp. 219. HERMANS, A. (1999). Hydrogeo- logisch en hydrochemisch onderzoek van het duingebied te Wenduine - De Haan. M.Sc. Thesis, RUG, Gent, pp. 92. STUYFZAND, P.J. (1986). A new hydrochemical classification of watertypes: principles and application to the coastal dunes aquifer system of the Netherlands. Proceedings of the 9th Salt Water Intrusion Meeting, Delft, 641-655. VAN HAECKE, K. (1998). Hydrogeochemisch onderzoek naar de zoet-zoutwaterverdeling langs een profiel loodrecht op de kustlijn (Wenduine - De Haan). M.Sc. Thesis, RUG, Gent, pp. 98. WALRAEVENS, K. (1990). Hydrogeology and hydrochemistry of the Ledo- Paniselian semi-confined aquifer in East- and West-Flanders. Acad. Analecta 52, 3, 11-66.