The potential of seismic imaging in marine archaeological site … · identify archaeological artefacts lying on the seabed. However, the main disadvantage of those techniques is

R e lic ta 6 (2010), 219-236 219

The potent ia l of s e i s m ic imaging in marine arch a eo lo g ica l s i t e in v es t iga t ion s

Tine M issiaen1

IntroductionOver the last decades marine acoustic techniques have become a successful investigation tool in archaeological studies, often focused on nearshore areas. The most commonly applied techniques involve side-scan sonar and multibeam systems, aimed to identify archaeological artefacts lying on the seabed. However, the main disadvantage of those techniques is that they cannot penetrate the subsurface. The marine seismic method, due to its penetration capacity, allows the detection and imaging of submerged archaeological objects. The steady development towards very high and ultra-high resolutions2 has made the seismic technique a major asset in site-specific archaeological research.

One of the major applications concerns the study of (buried) wooden shipwrecks. Since the late eighties there has been a sharp increase in research of archaeological shipwrecks using marine seismic techniques3. W ith an ever-increasing num ber of wrecks and other marine archaeological structures being discovered, the necessity for management and preservation of underwater cultural heritage becomes increasingly im portant. Furthermore it is nowadays well recognized that buried artefacts are often best protected if left in situ4. Non-intrusive high-resolution acoustic techniques capable of locating and identifying wooden materials buried in the seabed, possibly also assessing their state of preservation or decay, are therefore becoming increasingly im portant.

The most recent international developments in marine archaeological studies increasingly focus on prehistoric archaeology, in particular the impact of hum an activities on submerged terrestrial landscapes5. Although undoubtedly many of these terrestrial sites may have been eroded or destroyed by rising sea levels, their existence and importance is nowadays generally accepted.

The reconstruction of the paleo-landscapes is not only an imp o rtan t requirem ent to help understand their archaeological potential, such as submerged or re-worked material from older deposits, but it may also provide key inform ation on the hum an evolution. Seismic techniques form an ideal tool to image these buried paleo-landscapes.

In Belgium, little attention has so far been paid to the potential of seismic techniques for marine archaeological studies - in stark contrast to The N etherlands and the u k where this research has gained considerable interest over the last years. Up to now, the main archaeological research on the Belgian Continental Shelf ( b es) has been focused on exposed shipwrecks, for example on the Vlakte van the Raan (‘t Vliegend Hert, Zeebrugge-site)6. An im portant challenge, however, lies in the detection and imaging of buried objects (such as shipwrecks or dykes) and the study of inundated terrestrial sites. Sea-levels were generally much lower in late Pleistocene and early Holocene times and only reached near-present levels during the later p art of the Holocene. The identification of buried tidal channel, dunes, lagoons, river valleys, reclaimed areas, etc should allow a better understanding of the archaeological potential of the b es , including the différent processes involved in reworking of archaeological artefacts.

Three different seismic case-studies will be discussed in this paper. The first case-study involves seismic imaging of a buried 17th-century shipwreck in the Dutch W adden Sea. The second case-study is located offshore Raversijde (Ostend, Belgium), and focuses on the possible remains of Roman dykes and salt and/or peat exploitation and the distribution of prehistoric tidal gullies. The th ird case-study finally focuses on an exposed shipwreck on the Buiten Ratel sandbank.

i Renard C entre o f M arine Geology, VakgroepGeologie en Bodem kunde, U niversiteit G ent,K rijgslaan28189,9000 G ent, België,tine.m issiaen@ ugent.be, Tel: +32 9 2644571.

2 Resolution is th e ab ility to separate tw o featu res th a t are very close together.3 B u l le r i . 1998; Q u in n eta l. 1997 & 2000.4 G arabello & Scovazzi (eds) 2003.

5 Lafferty eta l. 2006; W estley eta l. 2004.6 Xeebroek e ta l. 2006.

mailto:[email protected]

220 T . M IS S I A E N

SourceSource + receiver

receivers

s e a floor

layer

f i g . i S c h e m a tic d ia g r a m s h o w in g t h e p r in c ip le o f m a r in e s e is m ic im a g in g . A s e is m ic s o u rc e e m its a s o u n d p u ls e t h a t is r e f l e c te d a g a in s t

t h e s ea f lo o r a n d d e e p e r la y e rs ; t h e r e f le c te d s ig n a ls a r e r e c o r d e d b y a to w e d re c e iv e r . A s t h e b o a t m o v e s c o n t in u o u s ly , t h i s r e s u l t s i n a v e r t i

c a l im a g e a c r o s s t h e s e a b e d a lo n g th e s h ip ’s t r a je c to r y .

S c h e m a tisc h d ia g r a m van h e t p r in c ip e v a n m a r ie n e se ism isch e b e e ld v o rm in g . E e n a k o e s tisc h e b ro n z e n d t een g e lu id s g o l f u i t d ie w e e r k a a ts t w o r d t

op d e z e e b o d e m e n d e o n d e r lig g en d e lagen . H e t g e re fle c te e rd e s ig n a a l w o r d t o p g en o m en d o o r e e n o n tva n g er . D o o r d a t h e t m e e tsc h ip z ic h c o n tin u

v e r p la a ts t , w o r d t op d ie m a n ie r een v e r tic a le d o o rsn e d e va n d e o n d e rg r o n d b e k o m e n la n g s h e t tr a je c t va n d e boot.

1 M arine seismic imagingIn marine seismic imaging an acoustic source and receivers are towed behind a ship. The source emits an acoustic pulse that travels through the water and is reflected from the seabed and subsequent layers of the subsoil (fig. i). The reflected energy intensity depends on the different densities of the seabed and sub-seafloor layers, the denser (i.e. harder) the seabed or layer the stronger the reflected signal. The reflected signal then travels back through the water to the receiver (fig. i). The received signals are recorded, and as the ship constantly moves this will result in a vertical cross section through the seabed. So-called reflectors on the seismic image m ark the boundary between two distinct subsurface layers.

In order to image the shallow subsurface with the highest possible detail (and to allow the detection of small buried objects) a high vertical resolution is needed. This implies the use of high- frequency acoustic (seismic) sources, such as boomers7 and echo- sounders8. Boomer sources have a frequency ranging between 2 and 5 kHz, and offer a good compromise between resolution (20- 50 cm) and penetration (tens of meters up to hundred m or more). Echosounder sources are generally marked by a higher frequency (4 to io kHz), resulting in an increased resolution (10-20 cm) but often also a decrease in penetration depth9.

The ability to image buried wooden objects, such as wooden shipwrecks, will highly depend on the type of wood and the level of decay, the density of the surrounding sediment and the depth of the artefact. Studies in the UK have shown that oak and pine artefacts are acoustically detectable in a wide range of marine sediments, w ith sandy sediments generally giving a better contrast than sand-silt-clay m ixtures10. Increasing wood degradation, caused by bacteria, woodborers and (to a lesser extent) fungi, will often yield a better image. Burial in marine sediments, however, may lim it the progression of degradation due to a lim ited oxygen supply. W ooden artefacts exposed on the sea floor are usually well detected, but once the wood is heavily degraded seismic detection can become very difficult11.

The ability to image archaeological artefacts will also depend on the size and orientation of the object. Acoustic reflections from a shipwreck will mainly be caused by long boards. To allow detection the wavelength of the seismic signal should therefore not be much bigger than the size of the wooden boards. In our case, working w ith acoustic signal frequencies well over 3 kH z this should not be a problem12. The sediments surrounding a shipwreck are also often marked by scour features. In time these scour features may become filled in and buried. In some cases the di-

7 Electrom agnetic ally driven sound source 8 Type o f sound source where a pulse is genera- 9 M issiaen20o8.where a capacitor bank is discharged th rough a flat ted by passage o f a n electrical im pulse th rough a io A rn o tt eta l. 2005.coil. Eddy cu rren ts are generated in a lum in ium piezoelectric crystal. The crystal is physically de- 11 A rn o tt eta l. 2005.plates b enea th the coil, causing rapid repulsion o f form ed, generating a pressure wave. The transm it- 12 Frequencies > 3 kH z w ill yield a wavelength <the p lates and the creation o f a shock wave. tin g transducer also acts as a receiving transducer. 0.5 m (fo rV = 1500 m /s).

T he p o te n tia l o f se ism ic im ag ing in m a rin e archaeological s ite investig a tio n s 221

mensions of the scour features (often much larger than the wreck itself) may make them more easily recognized on seismic profiles. In the extreme case of a completely degraded wreck, no longer detectable by seismic imaging, or excavation of the wreck, the scour pits may form the only remaining evidence of the wreck site.

2 Case study 1: Seismic imaging of a buried wooden shipwreck, Wadden Sea

In July 2003 a seismic survey was carried out over a buried wooden shipwreck (‘Scheurrak SOi’) in the Dutch W adden Sea, 25 km NE of Den Helder. The wreck is roughly 25 m long and dates presumably from the late 16th cen tury13. From 1989 to 1997 the wreck has been excavated by divers of the racm (The Netherlands) - at tha t tim e the wreck was largely exposed on the sea floor (fig. 2). Results showed tha t the wreck had been broken lengthwise (the starboard side had broken off the hull of the ship)14. Since then, sedimentation has resulted in complete burial of the wreck under a th in layer of fine sandy silt. The sea floor in the wreck area is relatively flat, locally marked by large ripples and sand waves. The water depth varies between 6.5 and 8 m below m llw l (mean lowest low water level at spring tide).

In the fram ework of the im a g o project (acronym for “Inno vatief M eten aan Gezonken Objecten”) a dense network of 62

seismic profiles was recorded over the wreck site (fig. 3: left). Multibeam data, recorded simultaneously with the seismic data, indicate tha t the wreck is completely buried. O n these m ultibeam data we can also clearly observe the short poles, an external measuring grid placed by divers to mark the wreck area (fig. 3: right). Both 2D and 3D seismic data were recorded, the latter using an inflatable receiver array which was towed behind the vessel. The seismic source consisted of a 3.5 kH z echosounder. Average length of the seismic profiles was 200-250 m with a line spacing of roughly 7 m. Strong side currents made it not always possible to steer exactly along straight lines, as can be seen on fig. 3: left. Positioning was done using a short-range d g p s 15 system based on r t k 16 positioning which allowed to record real-time (x, y, z) positions each second with cm accuracy.

The quality of the seismic images was generally quite good, although locally some data were m arked by poor penetration which was likely due to the presence of small amounts of gas bubbles in the sediments. Fig. 4 shows an example of a 2D seismic profile and interpreted line-drawing across the buried wreck site (the location of the profile is shown in fig. 3). The profile crosses the wreck more or less obliquely and the wreck stands out clearly on the seismic image. Some deeper reflections observed on the left are most likely caused by the irregular edge of the westernmost wreck piece. The large diffraction (marked in red on fig. 4)

f i g . 2 S c h e m a tic v ie w s o f t h e s h ip w re c k ‘S c h e u r r a k S O i ’ i n t h e W a d d e n S e a , b a s e d o n a r c h a e o lo g ic a l e x c a v a t io n s . L e f t : T o p v ie w o f t h e

w re c k t h a t h a s b e e n b r o k e n i n tw o la rg e p ie c e s . R ig h t to p : 3D m o d e l o f t h e w re c k b a s e d o n d ra w in g s a n d th e m u l t ib e a m d a t a . T h e g re e n

p o le s w e re p u t i n a s a n e x te r n a l m e a s u r in g g r i d d u r in g e x c a v a t io n ( th e d iv e r m a r k s t h e s c a le ) . R ig h t b o t to m : s a m e 3D m o d e l o f t h e sea f lo o r

b a s e d o n m u l t ib e a m d a t a . T h e p a r t l y e x p o s e d r u d d e r c a n b e s e e n o n t h e f o r e g r o u n d . S p ik e s a r e d u e to t h e p a r t l y e x p o s e d p o le s . C o u r te s y

r a c m ( l e f t p i c tu r e ) a n d R A C M /R W S ( r i g h tp i c tu r e s ) .

I l l u s tr a t ie s va n h e t sch e e p sw ra k 'S c h e u r ra k S O i ’ in d e W a d d e n z e e g e b a se e rd op a rch eo lo g isch e o p g ra vin g en . L in k s : B o v e n a a n z i c h t va n b e t w ra k

d a t in tw e e g r o te s tu k k e n is g e b ro k e n . R e c h ts b o v e n : 3 D m o d e l v a n b e t w ra k , g e b a se e rd op d e te k e n in g e n en d e m u lt ib e a m d a ta .D e ge le m e e tp a a l-

tje s m a r k e r e n d e w r a k s ite (de d u ik e r d i e n t a is sch a a lm o d e l) . R e c h ts o n d e r : h e tz e l fd e 3 D m o d e l m a a r m e t in v u l l in g va n d e z e e b o d e m z o a ls g e z ie n

d o o r m u l t ib e a m d a ta . O p d e v o o rg r o n d is b e t g e d e e lte l i jk u i ts te k e n d e ro e r d u id e l i jk te z ie n . D e sch erp e s p ik e s in d e m u lt ib e a m d a ta z i j n vero o r

z a a k t d o o r u i ts te k e n d e p a a ltje s . M e t d a n k a a n r a c m ( l in k e r f ig u u r ) e n r a c m / r w s (rech te r fig u ren ) .

13 M anders2000.14 R ijksw aterstaat 2003.15 D ifferential Global Positioning System.

16 Real-Time K inem atic- position ing technique provides the real-tim e corrections resu lting in abased on the use o f carrier phase m easurem ents centim etre level o f accuracy,o fth e GPS signals where a single reference station


200

150

100

50

m e te rs meters

f i g . 3 L e f t : S e is m ic n e tw o r k ( t h i n b lu e l in e s ) r e c o r d e d o v e r t h e w re c k s i te i n t h e W a d d e n S e a . B la c k l in e s = w re c k c o n to u r s ( r e c o r d e d in

e x c a v a t io n w h e n t h e w re c k w a s s t i l l e x p o s e d ) . R ig h t : M u l t ib e a m r e c o r d in g o v e r t h e b u r i e d w re c k . T h e w re c k is n o w c o m p le te ly c o v e r e d b y

s e d im e n ts . B la c k l in e s = w re c k c o n to u r s . R e d d o ts = s h o r t p o le s u s e d b y d iv e rs i n 199 0 to u s e a s a n e x te r n a l m e a s u r i n g g r id . T h ic k b lu e l in e

= lo c a t io n o f t h e s e is m ic p r o f i le s h o w n i n fig . 4 . C o u r te s y r w s .

L in k s : O v e r z ic h t va n h e t se ism isch e n e tw e r k (d u n n e b la u w e l i jn e n ) d a t w e r d o p g en o m en o ver d e w r a k s ite in d e W a d d e n z e e . Z w a r te l i jn e n = co n

to u r e n van b e t w ra k , z o a ls w a a rg e n o m e n t i jd e n s d e o p g ra vin g en (to en b e t w ra k no g v r i j la g ). R ech ts : M u l t ib e a m o p n a m e o v er d e w ra k s ite . H e t

w ra k l ig t v o lle d ig b eg ra ven o n d e r e e n la a g s e d im e n t. Z w a r te l i jn e n = co n to u re n van b e t w ra k . R o d e s t ip p e n = u its te k e n d e p a a l t je s d ie d e s i te m a r

k e r e n .D ik k e b la u w e l i jn = lo c a tie v a n b e t se ism isch e p r o fie l b e sp ro k en in fig . 4 . M e t d a n k a a n r w s .

at a depth of 12 milliseconds (ms) below the sea floor is possibly related to the ship’s rudder (see also fig. 2). The data indicate that the deepest part of the wreck is covered by roughly 1.5 to 2 m of sediment. The sea floor surrounding the wreck is locally marked by sand waves w ith an amplitude of 50-70 cm and length up to 5 m. The seismic reflectors related to the wreck are m arked by a slight phase inversion17, which indicates a moderate state of decay of the wood. Due to the highly variable data quality (attributed to shallow gas), it was not possible to obtain a reliable 3D map of the wreck outline.

3 Case study 2: Seismic investigations off Raversijde

The Provincial Domain of Walraversijde, between Ostend and Middelkerke, has been the focus of a large-scale archaeological research project set up by the Flemish Heritage Institute (v io e ) and the province of W est-Flanders. The m ain archaeological findings include the rem nants of a Late Medieval settlement, both on the beach (see fig. 5: left) and more inland in the dune

area18. Due to severe coastal erosion the first settlement, dating from the late 13th century, was lost to the sea and relocated behind a dyke in the early 15th century.

In September 2005 part of a buried Roman dyke was discovered in Walraversijde. The dyke is over 11 m wide and a little over 1 m high, and has a total length of at least n o m 19. The dyke is mainly built of stacked clay blocks, on its western side reinforced with peat (fig. 5 : right). The dyke is oriented roughly perpendicular to the present coastline, which suggests that its purpose was most likely to em bank a tidal gully that stretched further inland20. In the 70’s traces of the Roman dyke have been found on the beach of Raversijde21.

Soil studies on land have indicated a highly complex Holocene history22. Locally, the coastal landscape has been heavily influenced by peat exploitation. This hum an impact is in some cases thought to be the origin of relief inversion. Peat-digging started already in Roman times and was revived in Late Medieval times (i5-i6th century). Various peat-digging pits have been described

17 A phase shift o f i8 o °o fth e signal waveform.18 Pieters 1992.

19 P ieters eta l. 2006.20 P ieters eta l. 2006.

21 Pieters 2007.22 Pieters e ta l. 1998.


san d w aves

. . ' . , , ^ ¿ 1 . . - ^ ; ------------

Rudder? 0 2m

f i g . 4 E x a m p le o f a 2D s e is m ic p ro f i le ( to p ) a n d i n t e r p r e t e d l in e - d r a w in g (b o t to m ) a c ro s s t h e w re c k s i te i n t h e W a d d e n S e a ( f o r lo c a t io n

o f t h e p ro f i le see fig . 3 ). D e p th b e lo w t h e w a te r s u r f a c e i n m i l l i s e c o n d s tw o -w a y t r a v e l t im e (2 m s = 1.5 m ) . T h e p ro f i le c ro s s e s t h e w re c k m o re

o r le ss o b l iq u e ly a n d th e w re c k o u t l in e is c le a r ly o b s e r v e d . T h e la rg e d i f f r a c t io n a t t h e b o t to m is p o s s ib ly r e l a t e d to t h e s h ip ’s r u d d e r .

V o o rb ee ld van een tw e e d im e n s io n a le v e r tic a le a k o e s tis c h e d o o rsn e d e (b o v e n ) e n g e ïn te r p r e te e r d e l i jn te k e n in g (o n d er) o ver h e t w ra k in d e W a d

d e n z e e (voor d e lo ca tie v a n h e t p r o fie l z ie f ig . 3 rech ts). D ie p te o n d e r h e t w a te ro p p e rv la k in m illis e c o n d e n (2 m s = z.5 m ). H e t p r o f ie l k r u i s t h e t w ra k

m in o f m e e r d w a rs e n d e c o n to u re n van h e t b e d o lv e n w r a k z i jn d u id e l i jk te z ie n . D e in h e t o o g s p r in g e n d e d i f fr a c t ie o n d e ra a n k o m t m o g e lijk o v er

een m e t h e t roer.

f i g . 5 L e ft: G r o u n d - p l a n o f a la te m e d ie v a l h o u s e o n th e b e a c h o f R a v e rs ijd e (A f te r C h o c q u e e l 1 9 5 0 ) . R ig h t : E x c a v a t io n o f t h e R o m a n

d y k e i n W a l ra v e r s i jd e . T h e d y k e is m a d e u p o f c a r e f u l ly s ta c k e d c la y b lo c k s . W e d g e - l ik e p e a t s o d s ( d a rk c o lo u r ) w e re u s e d to r e in f o r c e t h e

w e s te r n s id e o f t h e d y k e , i n t h e f o r e f r o n t a h o r i z o n ta l c r o s s - s e c t io n o f t h e s a m e p e a t s o d s . T h e l a t t e r c le a r ly in d i c a t e t h r e e s u c c e ss iv e r e i n

f o r c e m e n t p h a s e s o f th e d y k e .

L in k s : P la tte g r o n d v a n een la a tm id d e le e u w s h u is op h e t s t r a n d v an R a v e rs ijd e (n a a r C h o cq u ee l 19 $0). R e c h ts : O p g ra v in g va n de R o m e in s e d i jk

in W a lra v ers ijd e . D e d i jk is o p g eb o u w d u i t z o r g v u ld ig g e s ta p e ld e k le iz o d e n . W ig v o rm ig e v e e n b lo k k e n (d o n k e r b r u in e k le u r ) w erd e n a a n g e b ra c h t

te r v e r s te r k in g v a n d e w e s tk a n t va n d e d i jk ; op de v o o rg ro n d een b a s isd o o rsn e d e va n d e z e lfd e v e e n b lo k k e n . D e z e la a ts te w ijz e n d u id e l i jk op d r ie

o p een v o lg e n d e ‘b e w a p e n in g e n ’ va n d e d i jk .


in the area of Walraversijde, some of these pits have a diameter of over 14 m23.

In October 2007 a marine seismic reconnaissance survey was carried out offshore Raversijde. The scientific goal of this survey was multifold: (1) localisation and identification of fossil (prehistoric) tidal gullies; (2) detection of the old Roman dyke that was discovered nearby on land; ( 3) detection of old coastline and/or coastal defense features; and (4) detection of possible artefacts (including peat excavation) of the deserted late medieval village Walraversijde. This should eventually lead to a better understanding of the complex late-Holocene evolution and the human impact on this part of the Belgian coastal region.

During the seismic survey a parametric echosounder source was used. The latter emits two sound signals with a different frequency (100 kH z and 8-12 kHz)24. W hereas the high-frequency signal allows a very detailed image of the sea floor, the lower-frequency signal gives a detailed image of underlying structure. The echosounder was attached on a long iron pole fastened to the side of the ship (fig. 6). A motion sensor was used to filter out any movement caused by wave action. The fast pulse rate (25 pulses per second) and the low vessel speed (3-4 knots) resulted in a high lateral coverage (one data point every 5 to io cm). Positioning was done using a D G PS antenna with an accuracy of ±1 m.

In to tal 37 seismic profiles were recorded (22 profiles parallel to the shore, 15 profiles perpendicular to the shore) (fig. 7). The length of the profiles varies between 1.9 and 2.3 km, with a spacing of 100-200 m. A few short profiles were recorded at high tide in between the groynes and as close to the beach as possible. The quality o f the seismic images is generally very good. The profiles nearest to the shore are m arked by a large num ber of shallow reflectors and show a high level of detail in the sub-seafloor structure and stratification. Further out towards the open sea the quality of the images decreases. This is most likely due to an increase in biogenic gas in the shallow sedim ents which will absorb the seismic energy25. The sloping sea floor towards the shore is marked by a step-like form, creating d istinct terraces.

Figs. 8 to io show some examples of recorded seismic profiles and the interpreted line-drawings. The length and depth scale are respectively indicated below and left of the profile (depth conversion was done using a constant velocity of 1550 m/s). On the whole, a large num ber of fossil tidal channels were observed on the seismic profiles. In most cases the gullies were best seen on the N-S oriented profiles and on the nearshore E-W oriented profiles (up to 1 km from the shore). Due to the high spatial variability of the shallow sediments and the large profile spacing it was not always possible to accurately track the channel pattern, especially the minor gullies. Further offshore the seismic profiles were marked by less penetration but still various shallow parallel reflectors, in some cases sharply interrupted by shallow gas, could still be identified just below the sea floor.

No clear indications were found of the Roman dyke. But the chance is real tha t such a dyke (and also other old dykes) has been largely eroded by the strong floods and tidal currents that m ark this coastal area. A marked, more recent tidal gully could be observed clearly on all N-S oriented seismic profiles. The gully runs roughly parallel to the shoreline (yellow line on fig. 7). The latter seems to indicate that most likely it represents an artefact o f an old coastal defense feature built to dam a creek or tidal gully. The orientation of this gully, more or less perpendicular to the Roman dyke, seems to suggest that there is no link between the two. The exact age of this more recent gully is not known.

Seismic profile A (fig. 8)This profile is oriented NW-SE, perpendicular to the shore. The water depth ranges from 9 m in the N W to 5 m in the SE. Except for a shallowing towards the shore the sea floor does not show any marked morphological features. In a few places the sea floor reflector is interrupted; this is due to some m inor technical malfunction of the acoustic equipment. We can clearly identify different fossil tidal channels (marked by striped black lines). The age of these channels is not known due to a lack of core data, but a possible relation to the so-called Walraeve’s Yde gully is not unlikely. Although the profile crosses the presumed continuation of the Roman dyke, no clear traces of the latter were observed. The most recent tidal gully, marked in yellow, is likely a rem nant of an old coastal defense feature (yellow line on fig. 7).

Seismic profile B (fig. q)This profile is oriented SW-NE, parallel to the shore. It is located roughly 1.2 km from the shore in a water depth of roughly 9.5 m. Although the acoustic penetration seems to be locally decreasing we can still clearly observe a large fossil tidal channel (marked by the striped black line). The parallel infilling structure inside the gully also stands out visibly. Again, the age of the channel is not known. On the left of the profile, towards the west, an object can be seen protruding from the sea floor. Below the object the acoustic image disappears completely. Possibly we are dealing w ith a small shipwreck. The seabed near the wreck shows a m arked depression, probably caused by scouring (from tidal currents) around the wreck.

Seismic profile C (fig. io)This profile is oriented SW-NE, parallel to the shore and w ithin the low-tide mark. Although the water is very shallow (roughly 2.5 m deep), the seafloor echo is not a prominent feature, which indicates soft upper sediments. Various horizontal, parallel reflectors can be observed down to a depth of 4 m below the seabed. There are no indications of fossil tidal channels. The most striking features are a num ber of strong, short and often jagged reflectors in the upper 1-1.5 meters (marked by thick black arrows). Their shallow depth and rugged form suggest a man-made origin, possibly related to artefacts from the fishing village Walraversijde or extensive peat-exploitation. Due to the large profile spacing it is not possible to map the exact distribution of these shallow reflectors.

23 P ieters 1992. 24 W underlich e ta l. 2005. 25 M issiaenei#Z.2002.


f i g . 6 L e ft : P a r a m e t r i c e c h o s o u n d e r a t t a c h e d to a p o le o n th e s id e o f th e s h ip . D u r in g m e a s u r e m e n ts a t s ea th e e c h o s o u n d e r is lo w e re d

2 0 -3 0 c m b e lo w th e w a te r s u r f a c e . R ig h t : F is h in g v e ss e l u s e d f o r t h e s e is m ic s u rv e y o f f R a v e rs ijd e .

L in k s : P a r a m e tr is c h e e c h o s o u n d e r b ro n v a s tg e m a a k t a a n een p a a l a a n d e z i j k a n t v a n h e t m e e ts c h ip . T i jd e n s d e m e tin g e n w o r d t d e ech o so u n

d e r n e e r g e la te n to t z o ’n 20-30 cm o n d e r h e t w a te r o p p e r v la k . R e c h ts : H e t m e e ts c h ip d a t w e r d g e b r u ik t v o o r d e s e is m is c h e s u r v e y v o o r d e k u s t

v a n R a v e r s ijd e .

s R o m a n d yk e

f i g . 7 S e is m ic n e tw o r k ( t h i n b lu e l in e s ) r e c o r d e d o f f s h o re R a v e rs ijd e ( b a c k g r o u n d m a p G o o g le E a r th ™ im a g e ) . T h ic k b lu e l in e s = lo c a

t i o n o f p r o f i le s A -C s h o w n i n fig s . 8 to io . R e d l in e = l o c a t io n o f t h e R o m a n d y k e o n la n d ; th e s t r i p e d t h i n r e d l in e m a r k s i t s p r e s u m e d c o n

t i n u a t i o n a t s e a . Y e llo w l in e = r e c e n t t i d a l g u l ly d i s c u s s e d in fig . 9 . G r e e n l in e s = v a r io u s p r e h i s to r i c t i d a l g u l l i e s . B la c k l in e s = g ro y n e s .

O v e r z ic h t va n h e t se ism isch e n e tw e r k (d u n n e b la u w e l i jn e n ) d a t w e r d o pgenom en v o o r d e k u s t va n R a v e rs ijd e (a c h te r g r o n d k a a r t G oogle E a r th ™

a fb e e ld in g ) . D ik k e b la u w e l i jn e n = lo ca tie van p r o fie le n A - C b esp ro ken in fig . 8 t .e .m . io . R o d e lijn = l ig g in g v a n d e R o m e in s e d i j k op la n d ;

d e g e s tre e p te d u n n e ro d e lijn is een m o g e lijk e v o o r t z e t t in g va n d e d i jk op zee . G ele l i jn = rec e n te g e ti jd e n g e u l , b e sp ro ken in fig . 9 . G ro en e l i jn e n =

o verige (o u d e )g e ti jd e n g e u le n . Z w a r te l i jn e n = s tr a n d h o o fd e n .


O.d H

M i a

d a t a g a p ------------------

r y * . ̂

■ 1

-

' A

. ‘ ‘ -s.1

I 1 ■ ■ = = »

recen t tida l gullyfossil tid a l gullies

f i g . 8 S e is m ic p ro f i le A o f f s h o re R a v e rs ijd e ( f o r l o c a t io n s ee f ig . 7 ) . T h e i n t e r p r e t e d l in e - d r a w in g is s h o w n b e lo w . D e p th b e lo w t h e w a te r

s u r fa c e is in m e te r s . F o r d e s c r ip t io n o f t h e p r o f i le s ee te x t .

S e ism isc h p r o fie l A v o o r d e k u s t van R a v e rs ijd e (voor d e lo c a tie va n h e tp r o f ie l z i e fig . 7). D e g e ïn te r p r e te e r d e l i jn te k e n in g tv o r d t d a a r o n d e r g e

to o n d . D ie p te o n d e r h e t w a te ro p p e rv la k in m e te r . E e n g e d e ta il le e r d e b e sp r e k in g v a n h e t p r o fie l w o r d t gegeven in d e te k s t .

ES .. . . . . . . .. a — .

w reek? se a floor

£ 1 2Q

fossii tida l gullies

14

f i g . 9 S e is m ic p ro f i le B o f f s h o re R a v e rs ijd e ( f o r lo c a t io n s ee f ig . 7 ) . T h e i n t e r p r e t e d l i n e - d r a w in g i s s h o w n b e lo w . D e p th b e lo w t h e w a te r

s u r f a c e is in m e te r s . F o r d e s c r ip t io n o f t h e p r o f i le s ee te x t .

S e ism isch p r o f ie l B v o o r d e k u s t van R a v e r s ijd e (v o o r d e lo c a tie v an h e t p r o f ie l z i e fig . 7). D e g e ïn te r p r e te e r d e l i jn te k e n in g w o r d t d a a r o n d e r g e

to o n d . D ie p te o n d e r h e t w a te ro p p e rv la k in m e te r . E en g e d e ta il le e r d e b e sp r e k in g va n h e t p r o f ie l w o r d t gegeven in de te k s t.


flJa

C l0)Q

f i g . i o S e is m ic p ro f i le C o f f s h o re R a v e rs ijd e ( f o r lo c a t io n s ee f ig . 7 ) . T h e i n t e r p r e t e d l in e - d r a w in g is s h o w n b e lo w . D e p th b e lo w th e w a te r

s u r f a c e is i n m e te r s . F o r d e s c r ip t io n o f t h e p ro f i le s ee t e x t .

S e ism isc h p r o f ie l C v o o r d e k u s t va n R a v e rs ijd e (v o o r d e lo c a tie va n h e t p r o fie l z ie f ig . 7). D e g e ïn te rp r e te e r d e l i jn te k e n in g w o r d t d a a r o n d e r g e

to o n d . D ie p te o n d e r h e t w a te ro p p e rv la k in m e te r . E e n g e d e ta il le e r d e b e sp r e k in g van h e t p r o fie l w o r d t gegeven in d e te k s t .

4 Case study 3: Seismic imaging of an exposed shipwreck, Buiten Ratel sandbank

The th ird example involves an 18th-century wooden shipwreck on the Buiten Ratel sandbank, 9 miles (16.5 km) offshore from N ieuw poort. The wreck, which has a length of roughly 18 m, was exam ined extensively by divers between 1996 and 2005. A rtefacts recovered from the shipwreck, including ceram ics, weapons, tools, utensils, clocks and wood, bear a large resemblance to objects recovered from 18th-century voc-sh ips (Verenigde Oost-Indische Compagnie - Dutch East India Company)26. Up to now the identity of the shipwreck remains unknown. In 2007 a m ultibeam survey was carried out over the wreck site by the Flemish Hydrography (Vlaamse Hydrografische Dienst27). The multibeam image shows that the wreck has been broken in two pieces; the partly exposed anchor is clearly visible (fig. 11).

In October 2007 a seismic survey was carried out over the Buiten Ratel shipwreck. During the survey a param etric echosounder source was used, combined with a motion sensor. More information on the seismic source is given in section 3 and in Zeebroek et al. (this volume). Positioning was done using a d g p s antenna with an accuracy of roughly 1 m. In total 44 short profiles were recorded, crossing the wreck at different angles (fig. 12). The length of the profiles ranges between 500 and 1000 m. Due to strong tidal currents it was not always possible to sail straight lines.

Figs. 13 to 15 show a few examples of seismic profiles recorded over the Buiten Ratel wreck. The upper and middle pictures respectively show the low-frequency (io kHz) and high-frequency (100 kHz) seismic image, the lower picture shows an interpreted line-drawing. In general the wreck outline (thin red-yellow line) is observed more sharply on the high-frequency data. The low- frequency data are marked by deeper penetration into the sediments and therefore allow identification of buried objects.

On the seismic images we can clearly distinguish the wreck outline (bright red or yellow reflectors) and the sediment cover that spreads over the wreck surface (blue or blue-green reflectors). The wreck is entirely covered by sediments. The thickness of this sediment cover ranges from a few cm up to 50 cm or more. The low amplitudes of the sediment cover, compared with the high- amplitude seafloor reflection, suggests that we are dealing with relatively soft cover sediments (e.g. increased silt and/or mud content) in contrast to the surrounding sandy sea floor.

Although the different wreck parts stand out sharply on various profiles (e.g. fig. 14), the distinction is not always clear when we only look at the sediment cover. Most likely local sedimentation, induced by tidal currents, has partly silted up the gap between the wreck parts. The height o f the wreck above the floor is in general no more than 1 m. The depth of the bottom of the wreck remains unknown because the acoustic signals cannot penetrate

se a floor

26 Zeebroek e ta l. 2010(this volume).27 V laam se H ydrografie, Afdeling K ust van het A gentschap M aritiem e D ienstverlening en K ust ( m d k ) , V rijhavenstraat 3 , 8400 O ostende.


through the wreck. Only a few m inor buried objects were observed near or (partly) under the wreck on the seismic profiles. But possibly this gives a distorted view of the actual situation - the chance to identify such objects is relatively small taking into account the large spacing between the profiles.

Seismic profile 18 (fig. it)This profile is oriented roughly N-S. The wreck contours on the seismic image are quite disturbed, most likely due to side reflections caused by curved or p ro truding parts of the wreck. The sedim ent cover is less than ha lf a meter. The different wreck parts, well observed on the multibeam image, do not stand out clearly. Probably this is due to the fact tha t the profile mainly crosses over the northern edge of the eastern wreck part (see fig. 12: right). On the low-frequency profile (fig. 13: top) a buried object can be observed, most likely a small detached wreck piece. The th in blue lines on the low-frequency image (fig. 13: top) are probably due to acoustic noise related to the equipment.

Seismic profile 4 .2 (fig. ia.)This profile is oriented SW-NE and crosses bo th wreck parts. The wreck is covered by a th in layer of sediments, no more than a few dm thick. The different wreck parts can be clearly distinguished. No additional buried objects were observed near or underneath the wreck. The thin vertical lines on the low-frequency image (fig. 14: top) are probably due to acoustic noise related to the equipment.

Seismic profile to (fig. is)This profile is oriented W-E and only crosses the western wreck part. The surface of the wreck is quite irregular, and there is a clear difference in height between the two wreck parts. The seismic image agrees well with the multibeam data. The wreck is covered by a thin sediment layer no more than a few decimetres thick. The weak reflector below the sea floor most likely marks a small buried wreck piece. The wavy blue lines in the water column of the low-frequency image (fig. 15: top) are most likely related to fish.

f i g . l í M u l t ib e a m im a g e a c q u i r e d i n 2 0 0 7

o v e r t h e B u i te n R a te l w re c k s i te ( F le m is h

H y d ro g ra p h y ) . T h e tw o la rg e w re c k p ie c e s

c a n b e c le a r ly o b s e r v e d .

M u lt ib e a m a fb e e ld in g u i t 2 0 0 7 v an b e t w ra k

va n d e B u i te n R a te l (V la a m s e H y d ro g ra fisch e

D ie n s t) . Tw ee g r o te w r a k s tu k k e n z i jn d u id e l i jk

z ic h tb a a r .

anchor

- 9.6

- 9.8

- 10.0

- 10.2

- 10.4

- 10.6

- 10.8 £_c4->

-11 .0 œ"U- 11. 2

- 11.4

- 11.6

- 11.8

- 12.0


30

f i g . i 2 L e ft : S e is m ic n e tw o r k r e c o r d e d o v e r t h e w re c k s i te o n th e B u i te n R a te l s a n d b a n k , p l o t t e d a g a i n s t a m u l t ib e a m b a c k g r o u n d

(F le m is h H y d ro g ra p h y ) . R ig h t : C lo s e -u p o f th e m u l t ib e a m im a g e a n d s e is m ic n e tw o r k . T h ic k o ra n g e l in e s m a r k t h e lo c a t io n o f s e is m ic

p r o f i le s i8 , 4 2 a n d 30 s h o w n i n f ig s . 13 to 15.

L in k s : O v e r z ic h t va n h e t se ism isch e n e tw e r k d a t w e r d op g en o m en o ver h e t w r a k op d e B u i te n R a te l - z a n d b a n k , g e p ro je c te e rd op een m u lt ib e a m

a c h te r g r o n d k a a r t (V la a m s e H y d ro g ra fisch e D ie n s t) . R ec h ts : C lose-up va n h e t se ism isc h n e tw e r k en m u lt ib e a m k a a r t . D ik k e o ra n je l i jn e n m a r k e

ren d e lo c a tie v a n p r o fie le n 1 8 ,4 2 en 30 b esp ro ken in fig . 13 t .e .m . 15.


i a c o u s tic *n o s e

10 kHz

CLf10

8

9

10

h

125 m

13100 kHz

8

9

10

11

12

13

4!Tl

JL 3

se d im e n t cover

jLse a floor

w reck ou tline

~ V " " -: :/»--- s e a floor

— \

w reck p iece?

f i g . 13 S e is m ic p r o f i le 18 o v e r t h e B u i te n R a te l w re c k s i te ( f o r l o c a t io n see f ig . 12). T o p : L o w - f r e q u e n c y im a g e ( io k H z ) . M id d le : H ig h -

f r e q u e n c y im a g e (100 k H z ) . B o t to m : I n t e r p r e t e d l in e - d r a w in g . D e p th b e lo w th e w a te r s u r f a c e is i n m e te r s . F o r t h e d e s c r ip t i o n o f t h e p r o

f i le , see te x t .

S e ism isc h p r o fie l 1 8 over h e t w r a k op d e B u i te n R a te l z a n d b a n k (v o o r d e lo c a tie van h e t p r o fie l z ie f ig . 1 2 ) . B o v e n : L a a g fr e q u e n t b e e ld ( i o k h z ) .

M id d e n : H o o g fr e q u e n t b ee ld (1 0 0 k h z ) . O n d e r: G e ïn te rp r e te e r d e lijn te k e n in g . D ie p te o n d e r h e t w a te ro p p e rv la k in m e ter . E e n g e d e ta ille e r d e

b e sp r e k in g van h e t p r o f ie l w o r d t gegeven in d e te k s t.


■

10kH z

E IQ

• .J . ’ -71 *■■ ■ V2* •, •-,. 0>«V-u 11

5 m

100 kHz3

9

r i o

o_■Si

12

13

W Ised im en t cover

se a floor

j L \ r 61—w reck outline

....................................\ X -.......; ƒse a floor

d ifferen t w reck p a rts

f i g . 14 S e is m ic p ro f i le 4 2 o v e r t h e B u i te n R a te l w re c k s i te ( f o r lo c a t io n s e e f ig . 12). T o p : L o w - f r e q u e n c y im a g e ( io k H z ) . M id d le : H ig h - f r e

q u e n c y im a g e (100 k H z ) . B o t to m : I n t e r p r e t e d l in e - d r a w in g . D e p th b e lo w th e w a te r s u r f a c e in m e te r s . F o r d e s c r ip t io n o f t h e p ro f i le s ee t e x t .

S e ism isc h p r o f ie l 4 2 o ver h e t w r a k op d e B u i te n R a te l z a n d b a n k (voor d e lo c a tie va n h e t p r o fie l z ie f ig . 1 2 ) . B o v e n : L a a g fr e q u e n t b e e ld ( i o k h z ) .

M id d e n : H o o g fi 'e q u e n t b ee ld (1 0 0 k h z ) . O n d e r: G e ïn te rp r e te e r d e l i jn te k e n in g . D ie p te o n d e r h e t w a te ro p p e rv la k in m e ter . E e n g e d e ta ille e r d e

b e sp r e k in g va n h e t p r o f ie l w o r d t gegeven in d e te k s t.


3

9

10

Cl

-ë h

12

13

3

9

I io

Cl-S 11

12

13

W E

acco u stic noise J -

100 kHz

e

9 w reck outlinesed im en t coverso ft sed im en ts

10 se a floorse a floor

H

w reck p iece?125 m

13

T3

f i g . 15 S e is m ic p ro f i le 30 o v e r th e B u i te n R a te l w re c k s i te ( f o r lo c a t io n see fig . 12). T o p : L o w - f re q u e n c y im a g e ( io k H z ) . M id d le : H ig h - f r e

q u e n c y im a g e (100 k H z ) . B o t to m : I n t e r p r e t e d l in e - d r a w in g . D e p th b e lo w th e w a te r s u r f a c e i n m e te r s . F o r d e s c r ip t io n o f t h e p ro f i le s ee t e x t .

S e ism isc h p r o fie l 30 over h e t w r a k op d e B u i te n R a te l z a n d b a n k (v o o r d e lo ca tie va n h e t p r o fie l, z ie f ig . 12). B o v e n : L a a g fr e q u e n t b e e ld (io k h z ) .

M id d e n : H o o g fi 'e q u e n t b e e ld (100 k h z ) . O n d e r: G e ïn te rp r e te e r d e l i jn te k e n in g . D ie p te o n d e r h e t w a te ro p p e rv la k in m e ter . E e n g e d e ta il le e r d e

b e sp r e k in g van h e t p r o f ie l w o r d t gegeven in d e te k s t.


Sum m ary and conclusionsThe use of marine seismic techniques for archaeological studies is illustrated by three different case studies. A first case study involved a buried wooden shipwreck in the Dutch W adden Sea. A lthough the acoustic visibility of buried wooden objects will depend on various factors (a.o. related to the wood and the surrounding sediment), clear detection is mostly possible. In our case the différent wreck parts, covered by roughly 2 m of sediment, stood out markedly on the seismic profiles. The images confirm the existing inform ation regarding shape and setting of the wreck. The seismic data furtherm ore suggest a moderate state of decay of the wood.

The second case study focused on the nearshore zone off the coast of Raversijde. The seismic data allowed to identify a large num ber of fossil tidal channels. A more recent tidal gully, oriented roughly parallel to the shoreline, can most likely be linked to an old coastal defense feature. The exact age of this gully is not known. No offshore rem nants were found of the Roman dyke discovered behind the dunes in Walraversijde. The seismic profiles nearest to the beach show the indication of a large number of man-made features, possibly related to the fishing village or peat-exploitation.

The last case study involved an exposed 18th-century wooden shipwreck on the Buiten Ratel sandbank. The wreck has been broken up in two large pieces. The different wreck pieces are clearly distinguished on the seismic data. Recent sedimentation, possibly induced by tidal currents, has resulted in a thin cover of relatively soft sediments. The latter has also partly silted up the gap between the two wreck parts. The wreck sticks out less than i m above the sea floor. A few small buried objects were observed near and under the wreck; most likely these can be linked to detached wreck pieces.

The results from the different case studies clearly demonstrate the potential of very high resolution seismic imaging for marine archaeological site investigations, in particular buried objects. The challenge of these often shallow marine environments calls on advances in technology, together with a certain pragmatism and creativity. C areful survey design is needed using lightweight equipment and small, shallow draft boats. A high spatial survey accuracy, preferably in 3 dimensions, is essential. Line spacing should be adapted to the size and specifications of the site. For example in the Raversijde study a closer line spacing (in the order of meters or tens of meters) should allow to get a closer grip on the full extent and exact distribution of the fossil gully system and its possible relation to ancient coastal defence features. Detailed imaging of buried archaeological objects requires an even higher grid density. I f the location of the object is known this generally does not pose a major problem. The big challenge, however, is to find objects of which we do not know they are there.

It is likely that no single technique will provide all the answers. The inventive use of com plem entary m ethods, for instance merging marine seismic acquisition w ith ground-tru th coring and other geophysical measurements (e.g. electric or electromagnetic methods, land-based seismics) is likely the best strategy.

In the Raversijde study such added ground-truth data, taken on the beach at low tide, should result in improved detection and identification of the man-made features that were observed on the marine seismic data. Recent studies in the Verdronken land van Saeftinge have shown that such integrated approach is indispensable for a better understanding on the shallow sedimentary environment, and its archaeological potential28. This opens new perspectives for the use of geophysical techniques in marine archaeological site surveys.

Acknowledgem entsThe captain and crews of the ‘W estergat’ and ‘Last Freedom’ are gratefully acknowledged. The author would furtherm ore like to thank W im Versteeg, Koen De Rycker, Arne Bayens, Rem- co Romijn and Seger van den Brenk for their assistance during the data acquisition. The W adden Sea survey was carried out in the framework of the i m a g o project (‘Innovatief Meten aan Gezonken Objecten’) in collaboration with t n o - t p d , The Netherlands.

Sam envattingH et potentieel van seism ische beeldvorm ing in marien archeologisch onderzoekSeismische beeldvormingstechnieken worden de laatste jaren met groeiend succes toegepast in marien archeologisch onderzoek, onder andere voor scheepswrakken en bedolven paleo- landschappen. Vreemd genoeg werd hieraan in België to t nu toe bijna geen aandacht besteed. Een eerste voorzichtige stap werd gezet met twee verkennende surveys voor de kust van Raversijde en bij een scheepswrak op de Buiten Ratel zandbank. Aan de hand van deze twee voorbeelden én een eerdere studie van een begraven houten scheepswrak in de W addenzee (Nederland) illustreren we het potentieel van marien seismische metingen voor archeologische studies.

Bij seismische m etingen w ordt m et behulp van gereflecteerde geluidsgolven een beeld gevormd van de structuur van de ondergrond. Daartoe worden een akoestische bron en één of meerdere ontvangers achter een schip gesleept. De bron zendt geluidsgolven uit die weerkaatst worden op de zeebodem en de onderliggende lagen. Deze gereflecteerde signalen worden opgevangen door de ontvangers. Doordat het meetschip zich constant verplaatst wordt op die manier een verticale doorsnede van de zeebodem bekomen, langs het traject van het schip.

H et onderzoek op de Waddenzee, uitgevoerd in 2003 in het kader van het iMAGO-project, betrof een bedolven i6de-eeuws houten scheepswrak. H et w rak ligt begraven onder een sedi- mentlaag van maximum 2 meter dik, en is uiteengevallen in twee delen. Beide wrakdelen zijn duidelijk waarneembaar op de seismische beelden. De vorm en ligging van de wrakdelen bevestigen eerdere informatie van duikers (in dejaren 90 van de 20ste eeuw) toen het wrak nog gedeeltelijk aan de oppervlakte lag. De seismische data lijken te wijzen op een niet al te verregaande staat van ontbinding van het hout.

28 MissiaeneitfZ. 2008.


Op de seismische profielen voor de kust van Raversijde valt de aanwezigheid op van een aantal fossiele getijdengeulen. Door de grote afstand tussen de seismische lijnen (100 to t 200 m) is het niet steeds mogelijk om het precieze verloop van deze geulen te volgen. Duidelijke sporen van de Romeinse dijk, die aan land werd teruggevonden, zijn niet te zien. Opmerkelijk is echter een kleine subrecente getijdengeul die parallel loopt met de huidige kustlijn en die waarschijnlijk duidt op een oude kustverdediging. De grillige vorm van de ondiepe seismische data dicht bij het strand suggereert een antropogene oorsprong. Mogelijk hebben we te maken m et restanten van het eerste vissersdorp Walraversijde.

De seismische survey op de Buiten Ratel zandbank toont duidelijk aan dat het wrak begraven ligt onder een variabele sedi- m entlaag van een paar cm to t ruim een halve m eter dik. H et wrak steekt niet meer dan 1 m eter uit boven de zeebodem. De bovenkant van het wrak is vaak grillig, met een duidelijk hoogteverschil tussen beide wrakdelen. Op sommige plaatsen zijn de

wrakdelen gedeeltelijk aan elkaar gesedimenteerd. Een paar begraven objecten (mogelijk afgebroken wrakdelen) werden waargenomen vlakbij het wrak.

De resultaten van deze drie verschillende casestudy’s tonen duidelijk het nut aan van seismische opnames voor archeologisch onderzoek. H et vaak ondiepe water vergt een nauwkeurige planning, met gebruik van kleine bootjes en licht materiaal, en een lijnspatiëring die is aangepast aan de specificaties van elke vindplaats. Ondanks de veelbelovende resultaten kan één enkele techniek uiteraard nooit alle antwoorden leveren. Het inventieve gebruik van complementaire methodes is waarschijnlijk de beste strategie. Mariene seismiekkan bijvoorbeeld gecombineerd worden met ondiepe boringen of met andere geofysische metingen, zowel op zee ais aan land. Dat laatste is bijvoorbeeld van toepassing op Raversijde. Dit opent nieuwe perspectieven voor de ontwikkeling en het gebruik van seismische technieken voor marien archeologisch onderzoek.


Bibliographya r n o t t s., D i x j ., b e s t a . & g r e g o r y D . 2005: Imaging of buried archaeological materials: the reflection properties of archaeological wood, M arine Geophysical Researches 26,135-144.

b u l l j., Q u i n n R . & d i x j . 1998: Reflection coefficient calculation from marine high resolution seismic reflection (chirp) data and application to an archaeological case study, M arine Geophysical Researches 2 0 ,1-11.

C H O C Q U E E L a . 1950: Les civilisations préhistoriques et anciennes de la Flandre Occidentale d ’après l ’examen d ’objets leur ayant appartenu, Bruxelles.

g a r a b e l l o R . & S C O V A Z Z I t . (eds) 2 0 0 3 : The protection ofthe underwater cultural Heritage. Before and after the 2001 u n e s c o Convention, Leiden/B oston.

L A F F E R T Y B ., Q U IN N R . & b r e e n R . 2006: A side-scan and high-resolution Chirp sub-bottom profile study of the natural and anthropogenic sedim entary record of Lower Lough Erne, north western Ireland, Journal o f Archaeological Science 33, 756-766.

m a n d e r s M .R . 2000: The mysteries of a Baltic Trader. In: b e l t r a m e c . (ed.), Boats Ships and Shipyards. Proceedings of the 9th Int. Symposium on Boat and Ship Archaeology ( i s b s a 9), Venice 2000, Oxford 2003, 320-329.

M i s s i A E N T. 2008: Shallow M arine Acoustic Imaging in Environmental Investigations, Dissertation submitted for the degree of Doctor of Science, Geology, Ghent University.

M IS S I A E N T ., m u r p h y s., L O N C K E L . & h e n r i e t j . -P . 2002: Very high resolution seismic mapping of shallow gas in the Belgian coastal zone, Continental Shelf Research 22.16, 2291-2301.

M I S S I A E N T . , s l o b e . & D O N S E L A A R M . E . 2008: Comparing different shallow geophysical methods in a tidal estuary, Verdronken Land van Saeftinge, Western Scheldt, The Netherlands, Netherlands Journal o f Geology 87-2 ,151-164.

P i e t e r s M . 1992: Archeologisch onderzoek te Raversijde (stad Oostende, prov. West-Vlaanderen), Archeologie in Vlaanderen 11, 247-264.

P i e t e r s M . 2 0 0 7 : Onderzoek op het strand van raversijde, pionierswerk in de maritieme archeologie. In: G E V A E R T G. (red .), De archeologische site Raversijde-strand'. Academische zitting over maritieme archeologie, Raversijde, 6 november 2007, Brussel, 31-33.

P i e t e r s m ., B A E T E M A N c., D E M ID D E L E H . & E R V Y N C K A . 1998: The polder area of Raversijde (Ostend, Belgium) : a complex parent material affected by intense hum an interventions, Archaeology as a tool in geopedology, Proceedings ofthe 16th World Congress o f Soil Science, Montpellier, France 20-26/081998, w css/bdd/sym p. 16,1593-t ( c d ).

p i e t e r s m . , D E M E R R E I . & z e e b r o e k I . 2oo6: Dijk uit de Romeinse tijd aangesneden onder het middeleeuwse vissersdorp Walraversijde. In: b o s m a n a . , d e c l e r c q w . & h o e v e n b e r g

j . (red.), Romeinendag. Jaarlijks Belgisch congres voor Romeinse archeologie. Journée d ’archéologie Romaine. Conférence annuelle Belge d ’archéologie Romaine. Gent 0 6 - 0 S - 2 0 0 6 Gand, 93-97.

Q u i n n R ., b u l l j . & d i x j . 1997: Imaging wooden artefacts using chirp sources, Archaeological Prospection 4, 25-35.

Q u i n n R . , c o o p e r a . & w i l l i a m s b . 2 0 0 0 : M arine geophysical investigation of the inshore coastal waters of N orthern Ireland, International Journal o f Nautical Archaeology 2 9 .2 , 2 9 4 - 2 9 8 .

r i j k s w a t e r s t a a t 2003: Onderzoekslokaties i m a g o prijsvraag, Rijkswaterstaat Internal Report, november 2003, Lelystad.

W E S T L E Y k ., d i x j. & Q u i n n R . 2004: A Re-assessment ofthe Archaeological Potential ofContinental Shelves - Final Report. Aggregates Level Sustainability Fund Project 3362, English Heritage Report.

236 T . M IS S IA E N

w u n d e r l i c h j., w e n d t G. & M ü l l e r s. 2005: High-resolution echo-sounding and detection of embedded archaeological objects with nonlinear sub-bottom profilers, M arine Geophysical Researches 26,123-133.

z e e b r o e k i., D E M E R R E I . , L E N A E R T S T. & p i e t e r s M . 2oo6: De Scheldemonding vanuit (maritiem) archeologisch oogpunt. In: c o o s e n j . , m e e s j . , s e y s j . & f o c k e d e y n . (eds), Studiedag. De Vlakte van deRaan van onder het sto f gehaald, v l i z Special Publication 35, Oostende, 52-58.

Z E E B R O E K I . , P I E T E R S M ., A N D R É S -L A C U E V A C ., C A L U W É D ., D A V ID J . , D E F O R C E K .,

H A N E G A K ., L A M U E L A -R A V E N T Ó S R ., L E N A E R T S T ., M E D IN A R E M Ó N A ., M E E S F.,

M IS S IA E N T ., M U Y L A E R T L . , OP D E B E E C K E ., S T R E E L M ., V A N D E N H A U T E P ., V A N H E E S M .

& W A U T E R S E . 2010: Een i8de-eeuwse wraksite opdeBuiten Ratel-zandbank(Belgische territoriale wateren) (I): m ultidisciplinair onderzoek van het vondstenm ateriaal, Relicta. Archeologie, Monumenten- & Landschapsonderzoek in Vlaanderen 6, 237-327.

Documents

The potential of seismic imaging in marine archaeological site … · identify archaeological artefacts lying on the seabed. However, the main disadvantage of those techniques is