This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Contents lists available at SciVerse ScienceDirect
Dendrochronologia
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . d e / d e n d r o
Original article
Precise tree-ring dating of building activities despite the absence of bark:A case-study on medieval church roofs in Damme, Belgium
Kristof Haneca∗, Vincent Debonne
Flanders Heritage Agency, Koning Albert II-laan 19, 1210 Brussels, Belgium
a r t i c l e i n f o
Article history:Received 8 February 2011
Accepted 7 June 2011
Keywords:
Wooden cultural heritage
Dendrochronology
Historical buildings
Roof constructions
Timber transport
Bayesian model
a b s t r a c t
A detailed dendrochronological survey was performed on the medieval roofs of the Church of Our Lady(CoOL) in Damme, Belgium. Seen its complex architectural history, special attention was paid to the
identification of consecutive building phases, based on combined architectural historical research and
tree-ring dating. In total 64 increment cores were taken throughout the roof structures of the CoOL. All
rooftimbersare made ofEuropeanoak (Quercus robur / petraea), ofwhichonly fewhavesurviving sapwood
or bark. Tree-ring dating confirms the late 13th/early 14th century construction date of the roofs. For all
chronologies that were composed, the highest correlation values are found with reference chronologies
covering the catchment area of the river Meuse. From the dating results of the timbers of the CoOL it
becomes clear that the same timber source was used for nearly a century. On several of the examined
roof timbers, rafting joints were observed, demonstrating that the timbers were indeed tied together as
a raft and floated down the river.
By implementing sapwood estimates in a Bayesian chronological model (OxCal), tree-ring series with
surviving sapwood from coeval roof structures were combined in order to narrow down the time range
for the felling date. Based on the refined interpretation of the felling dates, several consecutive building
phases can nowbe identified and dated,leading to a newinterpretation of the architectural history of the
CoOL. Intriguingly, a marked interruption in building activities is observed around 1300. Probably this is
related to the instable politicalsituation at thattime,causedby the armed conflict that emerged betweenthe Count of Flanders and the king of France. Since Damme served as the outport of the riotous city
of Bruges, it was alternately seized by the French and Flemish, both consuming considerable amounts of
timberand other building materialsfor military fortifications. Potentially this ledto a shortage in building
materials and provoked a stop in building activities.
This paper demonstrates the power of Bayesian models to refine the interpretation of dendrochrono-
logical dates in architectural analyses of medieval historical buildings.
K. Haneca, V. Debonne / Dendrochronologia 30 (2012) 23–34 25
Fig. 3. Simplified plan of the roofs of the Church of Our Lady in Damme, with the exact location of sampling points for the dendrochronological analyses, carpenter marksand rafting joints. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of the article.)
target for a combined dendrochronological and architectural his-
torical study in order to unravel and date the historical building
activities.
Sampling strategy
The sampling strategy was guided by the simultaneous study of
architectural features, such as the typology of the roof structures
and visible changes in the layout of the masonry. Furthermore,
special attention was paid to the recording of all carpenter marks
visible on the rooftimbers.The layout of thesecarpenter marks pro-
vides valuable information about the planning and construction of
the roofs. Ascending series of numbered rafters are an indication
that a roof was installed in one campaign, whereas a scission in the
carpenter marks could point towards an interruption in the build-
ing activities or a later replacement of timbers. In the case where
carpenter marks have a randomorder, the reuse of older (roof)tim-
bersshould be considered when interpreting dating results derived
thereof.
The entire roof structure was built with oak timbers (Quercus
robur L. or Q. petraea (Matt.) Liebl.). Timbers with bark or partially
preserved sapwood were targeted for the dendrochronological
analysis.For eachassumedconstruction phase or typological group,
we aimed at taking at least eight increment cores. All cores were
Significant dating results for all chronologies from the CoOL. The agreement between the chronologies is expressed by t BP-values (Baillie and Pilcher, 1973) and the percenta
K. Haneca, V. Debonne / Dendrochronologia 30 (2012) 23–34 29
Fig. 5. 3D-representation of the roof above the two most westward bays of the central choir, with indication of reused, older rafters (brown). Scale bar represents 5m. (For
interpretation of the references to color in this figure caption, the reader is referred to the web version of the article.)
where a = 2.813579, and the residual standard deviation
= 0.416208.In order to obtaina confidence interval forthe probable number
of sapwood rings, the region with the highest probability density
is calculated. This is the shortest range that includes the desired
percentage of the probability in the probability density function
P (Bronk Ramsey, 2009). It should be clear that this method devi-ates from the approach to obtain the 95.4% confidence intervals
for a normal distribution where two times the standard deviation
fromthe meanprovides the boundaries for thisconfidence interval.
Fig. 6. Bar graph of the dating results for each chronology. The bars represent the length of chronologies, with their height being proportional to the number of tree-ring
series included in the chronology. The black parts represent the number of measured sapwood rings. The grey parts of the bars are an indication for the minimum number
of sapwood rings that are to be expected, when less than 8 sapwood rings could be observed. The horizontal lines delineate the interval for the felling date.
K. Haneca, V. Debonne / Dendrochronologia 30 (2012) 23–34 31
Fig. 8. Graphicaloutput from OxCal (v.4.1.3), showing theprobability of each potentialfelling date forall serieswith survivingsapwood (grey), andthe posteriorprobability
of the combination of 4 series from the first phase of the southern choir and 13 series from the second phase of the southern choir, central choir and the entire northernchoir.
building campaign. During a second building campaign, at least 42
years later, a roof was constructed above the two western bays of
the southern choir (between 1283 and 1291 AD). After this point,
the dendrochronological evidence cannot provide more detailed
information aboutthe chronological order of the building activities
on the remaining roofs. The estimated range for the felling dates of
all chronologies are overlapping, and no logical chronological order
was suggestedby the rangesof thefellingdates. Theoutcomeof the
Bayesian analysis (Fig. 8) demonstrates that these roofs are indeed
coeval and that the felling date is situated between 1312 and 1315
AD.
The refined interpretation of the dating results of the roof
timbers from hall-choir of the CoOL now points towards an inter-
ruption in building activities between 1291 and 1312 AD. A
potential explanation for this is the armed conflict between the
Count of Flanders, Guy of Dampierre (◦1226–†1305 AD), and the
king of France, Philip the Fair (◦1268–†1314 AD), between 1297
and 1305 AD. As one of the main gateways to Flanders from over-
seas as well as the outport of the riotous city of Bruges, Damme
was alternately seized by the French and the Flemish, both forti-
fying the town with ditches, walls and gates. These military works
must have consumed a considerable amount of building materials,
transport could easily have been done by small ships navigat-
ing through the estuary, avoiding the open sea. Consequently the
time needed to transport timber along the Meuse, via Dordrecht,
towards Damme is probably not an important factor to be taken
into account when interpreting dendrochronological felling dates.
One factor that was not taken into account so far is the time
neededfor seasoningfreshly cuttimbers. In England andWales it is
known from the combined efforts of dendrochronological research
and the examination of written records that most timbers were
used in a roof construction between 1 and 3 years after felling
(Miles, 2006). In the town of Lübeck (Germany) it was observed
that some roof structures of historical buildings contained timbers
from different felling campaigns, separated by up to 33 months,
whileothersonlyhad timbers with onesingle felling date (Eckstein,
2007). The occurrence of mixed felling dates within one roof con-
struction suggests that construction was dependent of the amount
of wood in stock at local timber markets. However, from the den-
drochronological dataset from Flanders’ WCH or written sources
no evidence emerges to support this for construction timbers. Also
during this research project on the CoOL, only few timbers with
attached bark were observed, what makes it hard to interpret the
‘seasoning factor’. It is assumed that ‘green’ wood was used as
building timbers. Therefore, felling dates are currently interpreted
as the starting date for the construction of a roof. Written his-
torical documents could potentially provide more details on this
topic. Nevertheless, including a ‘stockpile factor’ into the Bayesianmodel is possible (Miles, 2006) and would slightly shift the felling
date ranges. In this particular case however, it would not influence
the interpretation of the chronological construction history of the
CoOL.
From the case-study in the town of Damme it is clear that the
absence of bark on construction timbers impedes to take den-
drochronological dating to its limits and provide felling dates up
to the season. Therefore it is clear that a dendrochronological anal-
ysis on roof timbers benefits from taking into account all recorded
sapwood information. The results demonstrate the value of the
Bayesian modelling in OxCal for testing whether timbers belong
to separate building campaigns. Furthermore, this approach allows
to narrow down the range of potential felling dates when multiple
series with surviving sapwood are available from one coeval con-struction phase. However, it should be noted that in cases where
series with a considerable number of sapwood rings are included,
this approach tends to extend the outer range of the confidence
interval towards a younger age (Miles, 2006). The sapwood esti-
mates used in this study were deduced from the original Hollstein
(1980) data, but any substantial dataset on sapwood numbers can
provide parameters for a sapwood model dedicated to a particular
region (Haneca et al., 2009, Table 1).
Data accessibility
All tree-ring series that were recorded, analysed and presented
in this paper were submitted to the DCCD repository in TRiDaS
format( Jansma et al., 2010). The data and meta-data are accessible
following http://dendro.dans.knaw.nl.
Acknowledgements
We are indebted to Jan Van den Bulcke (Laboratory of Wood
Technology, Ghent University) for writing a Matlab® script to pro-
duce bar graphs representing dendrochronological dating results
(Fig. 6). Furthermore we would like to thank our colleagues Kris
Vandevorst for taking excellent photographs, and Nele van Gemert
for enhancing the quality of the illustrations.
References
Bailiff, I.K.,2007. Methodological developments in the luminescence dating of brickfrom English late-medieval and post-medieval buildings. Archaeometry 49,827–851.
Baillie, M.G.L., Pilcher, J.R., 1973. A simple crossdating program for tree-ringresearch. Tree-Ring Bulletin 33, 7–14.
Bayes, T., 1763. Essay towards solving a problem in the doctrine of chances. Philo-sophical Transactions of the Royal Society of London 53, 370–418.
Bernard, V., 1998. L’homme, le bois et la forêt dans la France du Nord entre le
Mésolithique et le Haut Moyen-Age. BAR International Series 733, 1–190.Bernard, V., Epaud, F., Le Digol, Y., 2007. Le bois: de la fôret au chantier. In: Epaud,F. (Ed.), De la charpente romane à la charpente gothique en Normandie. Evo-lution des techniques et des structures de charpenterie aux XIIe-XIIIe siècles.Publications du CRAHM, Caen, pp. 9–46.
Bonde, N., Tyers, I., Wazny, T., 1997. Where does the timber come from? Den-drochronological evidence of the timber trade in Northern Europe. In: Sinclair,A., Slater, E.,Gowlett,J. (Eds.), ArchaeologicalSciences1995. OxbowMonograph64, Oxbow books Ltd., Oxford, pp. 201–204.
Debonne, V., Haneca, K., 2011. Baksteen en boomringen: een verfijndebouwchronologie van het hallenkoor van de Onze-Lieve-Vrouwkerk in Damme(prov. West-Vlaanderen). Relicta. Archeologie,Monumenten- en Landschapson-derzoek in Vlaanderen 7, 67–100.
deVries, D.J., 1994. Bouwen in de late Middeleeuwen: Stedelijke architectuurin hetvoormalige Over-en Nedersticht. Stichting Matrijs, Utrecht, 512 pp.
Devliegher, L., 1971. Damme. Kunstpatrimonium van West-Vlaanderen, no. 5,Bestendige deputatie van de Provincieraad van West-Vlaanderen, Brugge, 196pp.
Eckstein,D., Baillie, M.G.L., Egger, H., 1984.Dendrochronological dating, Handbooksfor Archaeologists, no. 2. European Science Foundation, Strasbourg, France, 55pp.
Eckstein, D., 2007. Human time in tree rings. Dendrochronologia 24, 53–60.Eckstein, D., Bauch, J., 1969. Beitrag zur Rationalisierung eines dendrochro-
nologischen Verfahrens und zur Analyse seiner Aussagesicherheit.Forstwissenschaftliches Centralblatt 88, 230–250.
Eckstein, D., Wröbel, S., 2007. Dendrochronological proof of origin of historic tim-ber - retrospect and perspectives, in: Haneca, K., Verheyden, A., Beeckman, H.,Gärtner, H., Helle, G., Schleser, G.H. (Eds.), TRACE–Tree Rings in Archaeology,Climatology and Ecology, vol. 5. Proceedings of the Dendrosymposium 2006,April 20th–22nd, Tervuren, Belgium. Schriften des Forschungszentrums Jülich,Reihe Umwelt 74, Forschungszentrum Jülich GmbH, Jülich, pp. 8–20.
Eeckhout, J., Houbrechts, D., 2002. Charpentes et caves médievales à Bruges. Cam-pagne d’analyses 2001–2002. Laboratoire de dendrochronologie, Université deLiège, Liège, 106 pp.
Eißing, T., 2009. Kirchendächer in Thüringen und dem südlichen Sachsen-Anhalt.Dendrochronologie– Flösserei – Konstruktion.ThüringischenLandesambtes fürDenkmalpflege und Archäologie, Erfurt, 230 pp.
Haneca, K.,Wazny, T.,Van Acker,J., Beeckman,H., 2005. ProvenancingBaltic timberfrom art historical objects: success and limitations. Journal of ArchaeologicalScience 32, 261–271.
Haneca, K., ˇ Cufar, K., Beeckman, H., 2009. Oaks, tree-rings and wooden culturalheritage: a review of the main characteristics and applications of oak den-drochronology in Europe. Journal of Archaeological Science 36, 1–11.
Heinemeier, J., Ringbom, A., Lindroos, A., Sveinbjornsdottir, A.E., 2010. SuccessfulAMS C-14 dating of non-hydraulic lime mortars from the medieval churches of the Aland Islands, Finland. Radiocarbon 52, 171–204.
Hillam, J., Morgan, R.A., Tyers, I., 1987. Sapwood estimates and the dating of shortring sequences. BAR International Series 333, 165–185.
Hoffsummer, P., 1995. Les charpentes de toiture en Wallonie, typologie et den-drochronologie(XIe-XIXe siècle). Etudes et documents, monumentset sites, no.
1, Ministère de le Région Wallonne. Division du patrimoine, Namur, 173 pp.
34 K. Haneca, V. Debonne / Dendrochronologia 30 (2012) 23–34
Hoffsummer, P., 2002. Les charpentes du XIe au XIX siècle. Typologie et évolutionen France du Nord et en Belgique. Cahiers du Patrimoine, no 62, Editions dupatrimoine. Ministère de la Culture et de la Communication, Paris, 376 pp.
Hoffsummer, P., 2007. The evolution of roofing in Northern France and Belgiumfrom the11thto the18thcenturyas revealed bydendrochronology,in: Haneca,K., Verheyden, A., Beeckman, H., Gärtner, H., Helle, G., Schleser, G.H. (Eds.),TRACE—TreeRings in Archaeology, Climatology and Ecology, vol.5. Proceedingsof the Dendrosymposium 2006, April 20th–22nd, Tervuren, Belgium. Schriftendes Forschungszentrum Jülich, Reihe Umwelt 74, Forschungszentrum JülichGmbH, Jülich, pp. 21–27.
Hollstein, E., 1980. Mitteleuropäische Eichenchronologie. Verlag Phillipp vonZabern, Mainz am Rhein, 273 pp.
Houbrechts, D., 2008. Le logis en pan-de-bois dans les villes du bassin de la Meusemoyenne. Dossier de la commission royale des monuments sites et fouilles, no.44, Commission Royale des Monuments sites et fouilles, Liège, 314 pp.
Hughes, M.K.,Milsom, S.J.,Leggett, P.A.,1981. Sapwood estimatesin the interpreta-tion of tree-ring dates. Journal of Archaeological Science 8, 381–390.
Jansma, E., Brewer, P.W., Zandhuis, I., 2010. TRiDaS 1.1: the tree-ring data standard.Dendrochronologia 28, 99–130.
Miles, D., 1997. The interpretation, presentation and use of tree-ring dates. Vernac-ular architecture 28, 40–56.
Miles, D., 2006.Refinements in the interpretationof tree-ringdates for oak buildingtimbers in England and Wales. Vernacular Architecture 37, 84–96.
Millard, A., 2002. A Bayesian approach to sapwood estimates and felling dates indendrochronology. Archaeometry 44, 137–143.
Rackham, O., 1982. The growing and transport of timber and underwood. In:McGrail, S. (Ed.), Woodworking Techniques Before A.D. 1500. BAR InternationalSeries, Oxford, pp. 199–217.
Sass, U., Eckstein, D., 1994. Preparation of large thin sections and surfaces of woodfor automatic image analysis. Holzforschung 48, 117–118.
Van Eenhooge, D., 2009. De middeleeuwse sporenkappen van de Onze-Lieve-Vrouwekerk in Brugge, M&L,Monumenten.Landschappen & Archeologie 28 (2),21–45.
van Prooie,L.A., 1990.De invoer vanRijns houtper vlot1650–1795.Economisch—ensociaal-historisch jaarboek 53, 30–79.
Wagenführ, R., 2007. Holzatlas, 6th ed. Fachbuchverlag Leipzig, Leipzig, 818 pp.Wazny, T., 2005. The origin, assortments and transport of Baltic timber: historic-
dendrochronological evidence,in: Vande Velde, C.,Beeckman, H.,Van Acker,J.,Verhaeghe, F. (Eds.), Constructing Wooden Images: proceedings of the sympo-sium on the organization of labour and working practices of late Gothic carvedaltarpieces in the Low Countries,Brussels 25–26 October 2002,VUB Press, Brus-sels, pp. 115–126.