(PDF) Developer-Led Excavations, Geoarchaelogy & Bayesian Modelled Chronology of a Guild House and the Main Street at Medieval O

Norwegian Archaeological Review ISSN: 0029-3652 (Print) 1502-7678 (Online) Journal homepage: www.tandfonline.com/journals/sarc20 Developer-Led Excavations, Geoarchaelogy & Bayesian Modelled Chronology of a Guild House and the Main Street at Medieval Odense Federica Sulas, Cristiano Nicosia, Kirstine Haase, Mikael Manøe Bjerregaard & Søren Munch Kristiansen To cite this article: Federica Sulas, Cristiano Nicosia, Kirstine Haase, Mikael Manøe Bjerregaard & Søren Munch Kristiansen (01 Jul 2025): Developer-Led Excavations, Geoarchaelogy & Bayesian Modelled Chronology of a Guild House and the Main Street at Medieval Odense, Norwegian Archaeological Review, DOI: 10.1080/00293652.2025.2515515 To link to this article: https://doi.org/10.1080/00293652.2025.2515515 © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. View supplementary material Published online: 01 Jul 2025. Submit your article to this journal Article views: 265 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=sarc20 Developer-Led Excavations, Geoarchaelogy & Bayesian Modelled Chronology of a Guild House and the Main Street at Medieval Odense FEDERICA SULAS , CRISTIANO NICOSIA , KIRSTINE HAASE , MIKAEL MANØE BJERREGAARD AND SØREN MUNCH KRISTIANSEN In Scandinavia, archaeological investigations have made much progress in detailing the characteristics of the earliest towns. Yet, understanding of urban stratigraphies remains limited, and this gap hampers our ability to define site formation processes and develop contextual palimpsests of town life at a formative time of urban development. This paper presents the results of a multi-scale analytical program applied to developer-led excavations of a medieval guild house and a paved street in the centre of Odense, Denmark, using soil micromorphology, trace elemental geochemistry (ICPMS) and radiocarbon-based Bayesian modelling. By integrating excavation and microstratigraphic data, our analysis shows that: (1) the use of space in the early medieval city was more dynamic than hitherto expected; (2) the guild house had separate spaces with floors that were maintained and re-laid several times; and (3) domestic waste was used to raise street levels in 1078–1162. This study also demonstrates the viability of integrating geoarchaeological methods, especially soil micromorphology and geochemistry, in developer-led excavations to aid and enhance the interpretation of complex urban stratigraphies. INTRODUCTION In Scandinavia, remains of medieval towns are usually sealed underneath modern-day cities and often affected by later infrastructure constructions such as basements and pipelines. Developer-led archaeological excavations follow the European (Valletta) Convention on the Protection of the Archaeological Heritage (Council of Europe 1992), which prioritizes preservation of archaeological remains in situ or, where this is not possible, full recording and removal. Present-day conditions make Federica Sulas, Department of Historical Studies, University of Gothenburg, Box 200, Gothenburg 40530, Sweden E-mail:

Cristiano Nicosia, Dipartimento di Geoscienze, Università degli Studi di Padova, Padua, Italy Kirstine Haase, Centre for Urban Network Evolutions, Aarhus University, Aarhus and Odense City Museums, Odense, Denmark Mikael Manøe Bjerregaard, Odense City Museums, Odense, Denmark Søren Munch Kristiansen, Department of Geoscience, Aarhus University, Aarhus, Denmark © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. 2 Federica Sulas et al. developer-led archaeological excavations in urban settings particularly challenging due to infrastructure development and the need for full documentation, where in situ preservation is not viable. Over the past decades, advances in field and laboratory methods have expanded the potential for recovery and interpretation, including sampling and analysis of microscopic records preserved in archaeological deposits. However, most of these methodological advances, such as soil micromorphology and geochemistry, are not widely applied in commercial and developerled archaeology in Scandinavia, despite several exemplar studies that demonstrate the versatility and feasibility of these methods in the region (e.g., Engelmark and Linderholm 2008, Linderholm et al. 2019, Linderholm 2021, Linderholm et al. 2023, Macphail et al. 2013, 2016, 2022, Sageidet 2000, 2013). Applications are nonetheless growing in Scandinavian and other European regions where soil micromorphology and geochemistry are increasingly applied in archaeological contexts in cities (e.g., Banerjea et al. 2016, Gebhardt and Langohr 1999, Milek 1997, 2012, Devos et al. 2020, Nicosia et al. 2012, Mazurek et al. 2016, Wouters et al. 2017, Shelley 2005, p. 178, Sulas et al. 2022, p. 175, Trant et al. 2024). In urban archaeology, field archaeologists excavate and record complex, fragmented deposits and contexts, determining the stratigraphic and chronological correlations between them. Challenges arise in interpreting these deposits and understanding fundamental aspects such as the nature and pace of formation and post-depositional processes. Are the deposits a result of human activities occurring over short periods of time? Or have they accumulated at a slow pace? What is the nature of these activities? Were these activities deliberate or unintentional? To answer these questions, fragmented deposits and site formation processes need to be characterised at multiple scales, capturing changing properties and features across space and over time, as shown by recent studies (e.g., Reid et al. 2023, Trant et al. 2024). Geoarchaeological and archaeological science methods, including soil micromorphology, geochemistry and Bayesian modelling of chronological data can provide high resolution details, records of features and processes in urban stratigraphy as shown, for example, in research on medieval European towns using soil micromorphology and, less frequently, geochemistry (e.g., Shahack-Gross et al. 2005, 2012, Dejmal and Brown 2014, Devos et al. 2013, 2020). In the Flanders (Belgium), for example, archaeological soil micromorphology solved the mystery of artificial hills (mounds/mottes) found in medieval towns by showing that they resulted from multiple phases of occupation, waste dumping and groundwater processes (Gebhardt and Langohr 1999). In these contexts, temperate climatic conditions offer both opportunities and challenges to examining past urban stratigraphies (see e.g., Kenward and Hall 2008). Factors such as temperature, soil moisture and reduction-oxidation (redox state) influence the preservation of organic and inorganic materials. In Scandinavia, archaeological excavations and material culture studies have made much progress in detailing the nature, development and characteristics of the earliest towns. A fast-growing strand of research is elucidating the ways in which settlements evolved into towns, the role of trade in such developments, social structures and the fabric of urban dwellers to mention but a few trending themes (e.g., Christophersen 2015, Petersén et al. 2015, Haase 2022). Yet, understanding of urban stratigraphies remains limited, and this gap hampers our ability to define site formation processes and develop contextual palimpsests of town life at a formative time of urban development in Scandinavia. Developer-Led Excavations In 2013–2016, developer-led excavations in the centre of the medieval town of Odense (site code OBM9776) have exposed portions of a street paved with gravel, dating to the second half of the 11th century, and remains of a guild house, dating to the late 12th to the late 14th century (Fig. 1, Haase 2017, Bjerregaard 2019). In medieval northern European towns, guild houses and main streets were spaces of hybrid and fluid activities, and movement (e.g., Christophersen 2015). Written sources indicate that guild houses were numerous but only a few have been recorded and excavated in Scandinavian medieval towns to date (see e.g., Brink 1996). Remains of guild houses are often difficult to identify archaeologically as they do not necessarily have specific features related to their function as guild houses. In Odense, the excavations were able to record and define the general layout of a guild house and the town’s main street, 3 but provided elusive records on the pace of the build-up and the activities that took place within these spaces. Key archaeological deposits within and outside the guild house and in the main street area were sampled for soil micromorphology, geochemistry and radiocarbon dating. Some of the analytical datasets have been presented in technical reports (Kristiansen et al. 2021, Sulas et al. 2021), but they have hitherto not been integrated into an overarching understanding of the site and the site formation processes. This paper combines excavation findings, soil micromorphology, geochemical and dating records: (1) to define the stratigraphy, use of space and post-depositional history of the guild house and the main street; (2) to expand the understanding of medieval Odense; and (3) to demonstrate how geoarchaeological and dating analyses can contribute to developer-led archaeology in urban settings. Fig. 1. Location and plan of central Odense, Denmark, and the excavation site OBM9776 The red shapes mark excavation trenches discussed in this paper; the blue outline show the area of previous investigations (OBM 9776 I/16-05-2013). Background maps: left, Historiskatlas.dk, 2016 Standard, and inset from Braun & Hogenberg, Civitates orbis terrarum, Otthoniae c. 1593; right, GeoDenmark data, SDFE (Agency for Data Supply and Efficiency 2022). 4 Federica Sulas et al. MATERIALS AND METHODS DEVELOPER-LED ARCHAEOLOGY AND TOWN EMERGENCE IN ODENSE Odense is first mentioned in 988 as an episcopal centre (Runge and Henriksen 2018, Christensen et al. 2019). Until recently, little was known about the early history of Odense during the Viking Age and the following centuries. Settlement remains dating to the 10th century were found on the southern floodplain of the river Odense Å where the (Trelleborg type) ring fortress of Nonnebakken was established at the end of the 10th century (Runge and Henriksen 2018). Whilst the fortress was likely in use only for a few decades, the medieval town developed on the northern side of the river. In this area, extensive town development in the 2010s led to large-scale, developer-led archaeological excavations and new research into the establishment and development of Odense. These excavations exposed settlement deposits spanning c. the 10th century to the present day, including remains of houses, wells, stables, craft production and other activity areas (Runge and Henriksen 2018, Haase 2019a, 2019b, 2022). The records show that the medieval town developed over an elongated and well-drained land north of Odense Å (Runge and Henriksen 2018, p. 5). Analysing soil micromorphological samples from archaeological deposits exposed west of the guild house (Fig. 1), Macphail et al. (2015) detected evidence of ground raising, using (alluvial) minerogenic turf, and occupation characterised by the disposal of hearth/kitchen waste and construction debris, including (secondary) phosphate features pointing to the presence of latrine waste. In some places, the turf soil material used for ground raising was mixed with highly phosphatised material and covered by clean sands. Remains of marine plant material indicated the import of this material for floor covering. At the top of the sequence, a sandy clay loam was covered by a lime plaster surface. ARCHAEOLOGICAL INVESTIGATIONS AT THE GUILD HOUSE AND THE MAIN STREET Developer-led excavations conducted in the centre of Odense in 2013–2016 exposed the remains of a street paved with gravel dating to the second half of the 11th century (Haase 2017, Bjerregaard 2019). The portion of the paved street excavated measures about 8 m in width and 50 m in length (Fig. 2). The street was the main thoroughfare of the town and was part of a grid of streets, which is thought to have marked a formalised and centralised manifestation of the structures of the early town. The oldest pavement was constructed directly on the subsoil, after removing topsoil and cultural layers associated with previous use of this area. The pavement was built using small fragments of flint (up to 10 cm) and variable concentrations of anthropic inclusions (twigs, Fig. 2. Excavation trenches at OBM9776, Odense, Denmark. Top: part of the paved main street with a section of the deposit on top; middle and bottom: sections in the guild house. Odense Byer museum Developer-Led Excavations wood, animal bone, antler) were embedded in this sandy matrix. The deposit appeared ‘compressed’ into a very compact surface. In some areas, wheel ruts were recorded during excavations. In the late 12th or early 13th century, the area was raised (c. 50 cm) using organic-rich material (Haase 2017, Bjerregaard 2019). On top of these layers, a series of small wooden market stalls were established, occupying the southern part of the street. In an adjacent plot, remains of the outer walls of a brick-built structure were excavated (Fig. 2). Remains of floors, a hearth, possible inner walls, and other deposits were recorded in the interior and would have been part of the ground floor, indicating that the whole structure measured approx. 11 × 7.7 m. Historical sources reveal that the building in this location functioned as a guild house in the 15th century (Christensen 1988). The guild house served as a place for gatherings for guild members on special occasions such as church holidays or funerals. Radiocarbon dates from the house deposits ranged from the mid-13th century to the early 17th century (Bjerregaard 2019). The house was demolished in the late 16th century as part of reorganisation in this part of the town (Christensen 1988). During the excavation, it had become clear that the guild house was erected on a plot of land that had previously been part of an adjacent cemetery since two graves were found below the house, one of which had been cut by the building’s foundation (Haase 2019a). SOIL SAMPLING AND ANALYTICAL METHODS Soil samples were collected during the excavation to test if and how geochemistry and soil micromorphology could shed light on the nature, activities and post-depositional history of the main street and guild house (Fig. 3). This was partly done for experimental purposes, as 5 these methods are not commonly applied in developer-led excavations in Denmark, but also to address specific research questions that emerged during the excavations (Table 1). For the guild house, the questions posed concerned which activities had taken place in the building and if there were any traces of passage in the various areas of the building. One question for the main street was whether the deposits on top of the pavement represented slow accumulation of dirt, or a ‘one-off’ event of a significant landfill. Multielement chemical analysis was performed on thirty-two soil bulk samples using Inductively Coupled Plasma Mass Spectrometry (ICPMS) to determine the concentration of thirty-five elements (4-acid extraction; full protocol and data are reported in Kristiansen et al. 2021). The geochemical samples are from archaeological layers and key deposits, documented in several excavation trenches (Fig. 3), and include twenty-three samples from the same contexts where seventeen micromorphology blocks were sampled, and six from soil monoliths (10 × 30 cm) also used for soil micromorphological analysis. For quality control, we also analysed duplicate samples (n = 20) and internal standards. Result screening excluded from the analysis data with low statistical correlation value (r < 0.95 Pearson correlation value), below detection level or associated with an element potentially volatile during sample processing (e.g., arsenic) or which had similar concentrations across the entire sample set. The analysis focused on the concentrations of ten elements that can reflect an anthropogenic signature (sensu Trant et al. 2024): silver (Ag), barium (Ba), calcium (Ca), copper (Cu), manganese (Mn), nickel (Ni), phosphorus (P), lead (Pb), tin (Sn) and strontium (Sr). To mitigate the lack of reference soil chemical data from the area, we compared the results from this study with the chemical properties of modern Danish 6 Federica Sulas et al. Fig. 3. Schematic plans showing the positions of soil samples (dots) for geochemical (ICPMS) and micromorphological analyses Developer-Led Excavations 7 Table 1. List of samples and contextual information. The sampling was conducted when trench excavations were ongoing and layer identification (and labelling) was underway. For this reason, a few individual samples are associated to different layers, and some individual layers are paired with more than one sample. GUILD HOUSE Context Eastern area section FCB Walkway FCH FCK Western area FCL Sample n Excavation layer Key question Function & activities OBM 1 (X10494) OBM 2 (X10508) OBM 18 (X1057 FCH)* OBM 3 (X10514) OBM 19 (X10513 FCK) OBM 4 (X10517) 1 3848 fill with brick rubble 5426 walkway 5381 sand floor, poss. bedding 5326 dirt layer over floor 5340 3819 limestone floor 3738 decomposed layer, brick rubble 3207 decomposed layer, brick rubble Traffic 1 3848 fill with brick rubble 2 5426 walkway with humus 5381 sand floor, poss. bedding 5340 sandy floor 5326 dirt layer over floor 5340 3832 clay floor 3207 decomposed layer, brick rubble Function & activities 1 5588 humus 3 5380 brick floor 5412 soot 5340 sandy floor 5326 dirt layer over floor 5340 3819 limestone floor 3207 decomposed layer, brick rubble Function & activities 1 5585 gravel levelling 3848 fill with brick rubble 5425 humus 3829 clay floor 3207 decomposed layer, brick rubble (continued ) 8 Federica Sulas et al. Table 1. (Continued ). GUILD HOUSE Context Entrance hall FCM Sample FLB FQD Excavation layer Key question Traffic OBM 5 (X10520) OBM 20 (X10519 FCM) MAIN STREET Eastern part section FCJ OBM 6 (X10578) OBM 7 (X10581 LAG 4942) OBM 8 (X10584 LAG 4941) Western part FQE OBM 10 (X11520 FRONT) FLB/FQE/ FQD n 1 3848 fill with brick rubble 2 5425 humus 3829 clay floor 3830/3207 decomposed, brick rubble 1 4942 top of cultural layer 1 4942 fill of pavement 1 4941 fill of pavement 1 4969 pavement and 6040 fill OBM 11 (X11523 FRONT) 1 Ditto OBM 16 (X11561 FRONT LAG 1 Ditto 4969/6040) OBM 17 (X11571 FRONT LAG 1 Ditto 4969/6040) OBM 13 (X11529 FRONT LAG 1 / 6040/6135) OBM 14 (X11533 FRONT LAG 1 / 6040) OBM 15 (X11541 FRONT LAG 1 6040 fill of pavement 2 5625 6040/5625) OBM 22 (X11593 LAG 5625/ 2 6040) OBM 23 (X11594 LAG 5625/ 6040) OBM 12 (X11526 FRONT LAG 1 6040 fill of pavement 6040/5809 topsoils (Reimann et al. 2014). However, these modern data are from agricultural grasslands (V. Ernstsen, GEUS, pers. comm) that, in Denmark, have seen additions of lime, marl and fertiliser through centuries of farming. Soil magnetic susceptibility was measured on all thirty-two samples, using a Bartington MS2b dual frequency sensor, to study the concentrations of naturally (biological) produced Accumulation rate Dunghill? Dunghill? Accumulation rate & activities Accumulation rate Accumulation rate and anthropogenic heated iron oxides with magnetic properties. Geochemical results were analysed using descriptive statistics and Principal Component Analysis (PCA) and all calculations were made in Python 3.6 (the Numpy, Pandas, scikit-learn and Matplotlib modules). The PC’s higher than PC2 did not reveal interpretable patterns, and neither Developer-Led Excavations did attempts to run PCA on subsets of the data; hence, these are not reported here. Twenty-three micromorphology samples from key archaeological deposits in various trenches (Fig. 3) were processed into thin sections at the McBurney Laboratory for Geoarchaeology, University of Cambridge (French and Rajkovaca 2015). The micromorphological description follows international standards for terminology and concepts (Bullock et al. 1985, Stoops 2003). Identification and interpretation follow guidelines from reference textbooks (Stoops et al. 2010, Nicosia and Stoops 2017, Macphail and Goldberg 2018) and case studies relevant to the context under examination. Full data are reported elsewhere (Sulas et al. 2021). RADIOCARBON DATING AND MODELLING Radiocarbon dating was performed on twelve samples from the guild house area and sixtythree samples from the main street area (Supplementary Table SA). Three dendrochronological samples and two coins from the main street area provided additional dates (Daly 2019). The coins were minted during the reigns of Oluf Hunger (1086–1095) and King Niels (1104–1134). The radiocarbon ages were obtained from the Poznan Laboratory (Goslar 2015) and converted into calendar years using the international terrestrial atmospheric calibration curve IntCal20 (Reimer et al. 2020). At the guild house, radiocarbon samples were taken from layers in stratigraphic relation with the eastern area (7 samples) and the western area (5) (Fig. 3). In the main street area, the radiocarbon (63) and dendrochronological (3) samples, and the coin come from layers in stratigraphic position. The layers excavated were then grouped into four broad types of activities based on finds and stratigraphy: (1) a row of postholes (labelled FTR) creating a fence (FTQ) and other activities prior to the establishment of the paved main street; (2) the paved street (FTO); (3) the build-up 9 of deposits on the paved street (FSF); and (4) market stalls that were established over the build-up layers (‘Market Stalls’). A Bayesian statistical model was created by using the stratigraphy, the coin and the dendrochronological dates as prior information (Supplementary Table SA). The model was constructed using OxCal 4.4 (Bronk Ramsey 2009) for the guild house and the main street. In both cases, OxCal command Sequence was employed to produce a multiphase model. A boundary was then added between each phase, and each phase then added to the model in stratigraphic order with a boundary between them (Supplementary Table SA). The samples in the phases were unordered and phase onsets (or terminations) were estimated using the added boundaries between phases. The Agreement Index (A) is an indicator of good agreement between the chronological data and the Bayesian model. An Agreement Index below 60% was used as an indication of a problematic sample or model (Bronk Ramsey 2009). RESULTS GEOCHEMISTRY OF ARCHAEOLOGICAL SEDIMENTS, FLOORS AND FILLS Geochemical analyses aimed at identifying potential chemical markers of activities in the excavation trenches (Fig. 3). The results show distinctive differences in the concentrations of specific elements between the archaeological samples from both the guild house and main street and the reference contexts (Table 2). Phosphorus (P) content in all samples is significantly higher than the content found in reference contexts (10–20× higher compared to contemporary Danish agricultural soils, Reimann et al. 2014; and up to 100× higher than the content found in loamy surface soils in Danish natural woodlands today, Kristiansen and Dalsgaard 2000). Lead (Pb) concentration exceeds the national soil quality criteria in most of the guild house 10 Federica Sulas et al. Table 2. Geochemistry of selected anthropogenic markers in the soil samples. In bold, values above the contemporary national soil quality criteria in Denmark (Miljøstyrelsen 2018). MDL: minimum detection limit. The entire dataset is reported in Kristiansen et al. (2021) and accessible at: https://download. museumodense.dk/files/publikationer/Rapport-11-Soil%20geochemistry%20in%20medieval% 20Odense.pdf (accessed 25 June 2024). Ca Sample Layer GUILD HOUSE X10495 3848 fill, Eastern area (EA) X10509 3848 fill, EA walkway X10515 5588, humus, EA walkway X10518 5585 gravel, Western area (WA) X10521 3848 fill, WA entrance hall MAIN STREET X10585 4941 fill of pavement X10579 4942 fill of pavement X10582 4942 fill of pavement X11512 4969 pavement X11513 4969 pavement X11514 4969 pavement X11515 4969 pavement X11516 4969 pavement X11521 4969 pavement MDL /Section P % 0.01 0.001 Ag Ba mg kg−1 0.020 1 Cu Mn Ni Pb 0.02 0.1 1 0.1 Sn 0.1 Sr 1 FCB 3.57 0.289 0.505 393 88.6 766 8.8 44.5 4.3 196 FCH 2.82 0.277 0.505 402 58.8 705 10.4 34.2 5.6 170 FCK 5.41 0.345 1.17 417 165 732 15.0 104 13.5 203 FCL 5.31 0.299 1.31 389 249 609 11.8 61.9 20.5 191 FCM 3.13 0.383 0.963 471 258 718 12.5 61.2 FCJ 3.15 0.299 2.19 390 121 691 6.1 42.5 FCJ 3.87 0.431 2.69 444 81.6 865 6.3 118 FCJ 3.53 0.351 1.42 415 47.5 858 6.2 22.0 88.3 149 FTO 1.34 0.442 0.655 476 15.6 626 5.2 13.6 0.9 162 FTO 1.15 0.409 0.628 436 13.3 474 4.6 21.9 0.6 149 FTO 1.73 0.454 1.81 486 26.4 819 5.4 13.3 0.9 164 FTO 2.03 0.414 1.16 440 81.1 920 5.9 41.0 2.9 132 FTO 2.96 0.377 6.33 488 866 6.9 148 4.2 162 FQE 3.30 0.344 2.45 472 858 6.4 1.7 157 572 47.3 84.2 752 177 2.6 144 2.9 166 (continued ) Table 2. X11576 X11578 X11543 X11528 X11522 X11525 X11527 X11530 X11534 X11535 X11542 X11577 X11579 X11531 11 Sr (Continued ). Ca X11524 Developer-Led Excavations 4969 pavement 4969 pavement 4969 pavement 5625 fill of pavement 5809 fill of pavement 6040 fill of pavement 6040 fill of pavement 6040 fill of pavement 6040 fill of pavement 6040 fill of pavement 6040 fill of pavement 6040 fill of pavement 6040 fill of pavement 6040 fill of pavement 6135 fill of pavement P Ag Ba Cu Mn Ni Pb Sn FQE 2.81 0.405 2.12 397 60.8 673 7.1 100 2.0 153 FTO 0.73 0.190 0.025 467 3.1 339 4.5 12.5 0.7 143 FTO 0.76 0.191 <0.020 422 2.6 504 4.4 11.8 0.7 147 FLB 4.79 0.475 1.78 526 77.1 990 8.8 39.5 1.7 197 FQD 3.19 0.299 6.40 390 66.6 600 7.6 39.8 5.5 145 FQE 3.76 0.279 1.45 388 42.3 457 6.9 31.0 1.6 148 FQE 3.55 0.354 1.71 417 56.1 722 7.2 101 1.5 157 FQD 4.05 0.349 1.12 409 77.0 729 7.9 23.0 2.4 163 FLB 3.62 0.250 1.10 379 42.1 611 6.3 31.2 1.9 148 FLB 4.53 0.463 3.48 444 87.5 829 7.9 133 2.2 178 FLB 4.13 0.404 1.66 461 75.0 799 7.3 36.8 1.9 177 FLB 3.40 0.426 1.19 490 55.8 559 8.3 33.7 2.5 172 FSF 3.41 0.334 8.01 418 82.7 741 6.9 132 1.9 162 FSF 3.09 0.423 0.884 460 91.0 1010 8.4 24.4 4.6 185 FLB 3.90 0.307 0.805 408 52.4 6.5 25.2 2.0 155 samples (4 out of 5) and a few (8 out 25) from the main street. Copper (Cu) and selenium (Sn) are the only other elements above the soil contamination criteria but in one sample from a fill layer in the guild house (entrance hall, sample X10521, layer 3848). The highest concentration of anthropogenic elements was detected in the street layer (sample X11516, layer 4969). In general, all the samples from the main street yielded concentrations of silver (Ag), barium (Ba), copper (Cu), manganese (Mn), phosphorus (P), lead (Pb), selenium (Sn) and strontium (Sr) higher than those detected in the guild house samples. The latter had enriched levels 819 of nickel (Ni) relative to the levels found in the main street samples. Principle Component Analysis (PCA) revealed correlations between different chemical elements and between elemental clusters and contexts (Fig. 4). Insoluble mineral elements (hafnium Hf, uranium U, tin Ti and zirconium Zr) are found in the lower part of the loading plot. At the top of PC1 (explaining c. 55% of the variation), potassium (Na) represents the dilution of the samples with quartz sand as well as organic material, and it is closely related to the natural soil parent material. This pattern (PC1) does not change notably when individual elements are 12 Federica Sulas et al. removed. PC2 (c. 17% of the variation) displays a loose cluster associated mainly with anthropologically influenced elements. The main street samples are enriched in lead (Pb) and distributed along the PC2, suggesting that the variation of naturally occurring elements is the main factor controlling PC1. In one street sample (X11516), high values of copper (Cu) and lead (Pb) (shown in the lowest part of PC2) reflects a stronger anthropogenic effect in this sample. In the PCA, the guild house samples group differently from the main street ones likely due to different content of clay – which controls PC1 on the loading plot (Fig. 4) – and higher contents of the anthropic markers (lead Pb, copper Cu and phosphorus P) (Fig. 4). The guild house samples from the eastern area, the western area and the entrance hall had similar concentrations and showed the strongest anthropic signature. The difference in magnetic susceptibility when measured with high and low frequency respectively, is an indicator of biologically produced iron (Fe)hydroxides. In the loadings plots (Fig. 4), this loading is at the opposite (right) side of the plot as iron (Fe) and non-anthropogenic elements, suggesting an anthropogenic origin of the magnetic iron minerals. ARCHAEOLOGICAL SOIL MICROMORPHOLOGY Analysis of thin sections from the guild house and the main street identified six main categories of materials suggestive of different accumulation patterns, uses and post-depositional processes. In most thin sections, sub-units were identified based on differences in composition and these were labelled using small case letters from bottom to top. At sub-unit level, the following materials were recorded: cess and latrine waste, anthropic waste from cleaning of hearths and/or cooking, mortar floor, calcareous fine matrix floor, beaten earth floor and peat-like sediments (Table 3). In some instances, the same material was observed in different deposits, suggesting potential associations between different spaces. In other cases, the same material was found deposited in different ways across different Fig. 4. Principal component analysis (PCA) of chemical measurements on samples from the guild house (guild) and the main street (numbered), Odense (ODM9776). For full chemical and magnetic susceptibility data see Kristiansen et al. (2021). Developer-Led Excavations 13 Table 3. Key materials identified in the thin sections from medieval Odense. EA = Eastern Area; WA = Western Area; small case letters denote soil microfabric types. Identification based on: Angelucci (2017, pp. 223–230), Cammas (2018), Canti (2017a, pp. 141–142); Canti 2017b, pp. 181–188), Milek (2012), Rentzel (2017, pp. 281–297), and Stoops et al. (2017, pp. 189–199). The full micromorphological dataset is reported in Sulas et al. (2021) and publicly accessible at: https://download.museumo dense.dk/files/publikationer/Rapport-10-Soil%20micromorpholgy%20in%20medieval%20Odense.pdf (accessed June 25, 2024). Materials Cess and latrine waste Waste from cleaning hearts/ cooking Floor types Key characteristics Plant-derived material at different stages of decomposition (Figs. 5a,b and 6c,d); occ. secondary phosphates (vivianite), fungal spores. Area GUILD HOUSE EA room fill WA room fill & Entrance Hall MAIN STREET Fills Mix of burnt plant residues, large wood GUILD HOUSE charcoal (hardwood and softwood), microcharcoal; burnt/unburnt shell and EA room fill & bone fragments (Figs. 5c,d and 6b); ceramic, brick, flint/chert inclusions. walkway WA Entrance Hall Mortar floor: Medium to very fine sandy GUILD loam, with variable content of silt HOUSE (Fig. 6e); occ. microcharcoal and ash; EA room fill parallel disposition of elongated components (charcoal, shell). Calcareous fine matrix floor: Fine to very GUILD fine sandy loam with calcitic groundmass HOUSE (Fig. 5e,f). WA room fill & Entrance Hall MAIN STREET Fills above/ below pavement Thin section and sub-unit OBM OBM OBM OBM 1a, OBM 3a 4a,c 5a, OBM 20.3c 10 OBM 1e,g; OBM 3b,d; OBM 19.1j,k; OBM 19.2 g OBM 2a,c; OBM 18.1a,d; OBM 18.2b OBM 5d; OBM 20.1e,f OBM 1f,h; OBM 3c,f OBM 4b,d OBM 5c, OBM 20.1 g,e OBM 16b; OBM 17b (continued ) 14 Federica Sulas et al. Table 3. Materials (Continued ). Key characteristics Area Earth floor: Sandy clay loam with rounded sand grains and massive (apedal) microstructutre (Fig. 5g,h). Peat-like sediments Heterogeneous mix of sand, decayed plant residues (wood, grasses), seeds, burnt bone and eggshell fragments (Figs. 5i, j and 7a,b) spaces. Distinctive micromorphological features indicate the nature and pace of processes in particular spaces (Fig. 5 and Table 4). For example, horizontal orientation of components (e.g., sand, plant residues) can be the result of compaction and compression by trampling (Matthews et al. 1997, Rentzel 2017). Whether the trampling is by people or animals cannot be ascertained via soil micromorphology. However, in some cases, traces of animal excrements were detected alongside features derived from trampling. Three main deposits were identified in the thin sections from the guild house and recur in the eastern and western areas, and the entrance hall. First, a deposit of cess and latrine waste appears to have accumulated in situ at the bottom of the eastern area. Here, there is indication of re-flooring using loamy material, likely to seal off the underlying waste. In the western area, the same waste is less compact and homogeneous, possibly because of dumping, rather than gradual, in situ accumulation. The presence of a cess waste was also recorded in thin sections from excavations in the adjacent plot (OBM GUILD HOUSE WA room fill & Entrance Hall MAIN STREET Fills above/ below pavement MAIN STREET Fills above/ below pavement Thin section and sub-unit OBM 16a; OBM 17a OBM 6–8, OBM 10–11, OBM 14, OBM 16–17, OBM 22.1–2, OBM 23.1–2 9776 I, Macphail et al. 2015) (Fig. 1). The thin sections from the guild house show cessrich sub-fabrics with common organic-rich, phosphatised material and secondary phosphates (Fig. 6c–d). Subsequently, anthropic waste (from cleaning of a hearth and/or cooking) is found accumulated in situ in the eastern area. Different types of floors were then established: a mortar floor in the eastern area (Fig. 6a–b), and calcareous fine floors in the western area (Fig. 6e). Within this general sequence, specific conditions and patterns were also observed. For example, in the thin section from the ‘walkaway’ in the eastern area (OBM2, Fig. 3), a very thin sub-unit of finely laminated very fine sand and silt is indicative of water runoff on the surface, either because at some point this space was unroofed or because water and sediments were washed down from the doorway during strong rainfall events. The paved main street was sampled for micromorphological analysis in the eastern and western sides to investigate the deposit beneath the pavement, the fills above it and later deposits (Fig. 7). Most thin sections show Developer-Led Excavations 15 Table 4. Summary list of the areas where micromorphological indicators of accumulation rate, trampling and post-depositional processes were recorded. Field interpretation Micromorphological evidence in situ acc. Area GUILD HOUSE Eastern area, room fill Brick rubble Unit 2 Eastern area, walkway Sand and clay floors Western area, room fill Gravel levelling, brick rubble and clay floor Western area, Brick rubble and clay floor Entrance hall Cultural layer and fill of pavement MAIN STREET Fills Beaten earth floor Ditch a thick deposit of organic-rich, anthropic waste mixed with peat-like sediments that has undergone waterlogging conditions. The greatest differences concern the depositional history of this deposit. A thick deposit of peat-like materials with anthropic inclusions appears to be the result of rapid accumulation. Organic matter is relatively well preserved compared to what can be observed elsewhere, likely because of post-depositional waterlogging (Figs 7a, c, and d). The presence of faecal material, including secondary phosphates, and traces of compression (Fig. 7a) suggest that this deposit originated from the dumping of domestic waste, with trampling occurring after sedimentation was complete. It is unclear whether this organic-rich, anthropic, peat-like sediment is the same as the local turf material identified by Macphail et al. (2015) in the adjacent plot and associated with groundrising. BAYESIAN CHRONOLOGY The modelling of radiocarbon dates produced three Bayesian statistical models (Fig. 8, and Supplementary Table SA). The models for the dumping Units 2 & 4 Units 1 & 3 Unit 1 Units 1 & 4 Unit 2 X trampling X Unit 1 Units 1 & 4 X X X eastern and western areas of the guild house show good model agreement (Amodel:98 and Amodel: 85; Supplementary Table SA). The guild house was erected in the period between 1198 and 1386 – much earlier than 1435, the date given in written sources (Christensen 1988, p. 127) and assumed to mark the erection of the guild house. The model for the main street shows a low but acceptable model agreement (Amodel:70). The poor agreement detected in six samples may be due to small sample sizes, or the origin of the sample material (intrusive or redeposited). However, removing these samples from the model does not change the dating of the phases significantly, though it increases the model agreement to Amodel > 100. According to the model (Amodel:70, see also Haase and Olsen 2021), the earliest activities recorded in the main street took place in 987–1033 (95.4% posterior probability) and the paving was laid out in 1056–1088 (95.4% posterior probability). The street was in use and re-surfaced in 1072–1112 (95.4% posterior probability). The micromorphological samples are mainly related to the paving phase (1056–1088) and the superseding activities (1072–1112). 16 Federica Sulas et al. Fig. 5. Micromorphological indications across contexts: a. and b. latrine waste with high phosphatic material rich in plant-derived fragments, Eastern Area, OBM 1a; c. sand-enriched in charcoal (ch), ceramic (cm) and bone fragments (b) in the anthropic waste layer from cleaning of hearths and accumulated in situ, Western Area, entrance hall, OBM 5d; d. sand-enriched with charcoal (ring-porous wood), ash and phosphatic material from reworking of anthropic waste dumped in, Eastern Area, walkway, OBM 18.1d; e. and f. fine loamy (calcareous) material with parallel orientation of charred and uncharred remains, resulting from compression (trampling?), Eastern Area, walkway, OBM 18.1f; g. and h. apedal microstructure suggesting a beaten earth floor, Main Street, OBM 16a; i. thin section scan of peat-like material from the Main Street, OMB 10; j. plant material (red arrows) in peat-like material from Main Street, OBM 8. Red scale bar: 1 mm. Developer-Led Excavations 17 Fig. 6. Micromorphology of the guild house. Top: Eastern Area OMB 19, monolith sample and thin section scans; a. contact (arrows) between a lens of very fine loamy, calcareous material from a thin reflooring or sand spread (top, sub-fabric j) over anthropic waste (bottom, sub-fabric h) intermixed with rubble material, chaotic organisation and open structure pointing to rapid accumulation (dumping), OBM 19.1; b. anthropic waste from hearth, with embedded brick/ceramic rubble, burnt (fish?) bone 18 Federica Sulas et al. DISCUSSION Developer-led excavations defined the layout and dating of the guild house and main street. The results of geochemical, micromorphological and Bayesian statistical analyses enabled characterising the nature of materials and activities taking place in these spaces and elucidate on site formation and post-depositional processes. In general, geochemical results show enriched levels of phosphorus (P), copper (Cu) and lead (Pb) in both the guild house and the main street samples. Phosphorus (P) enrichment is likely only slightly related to the extraction method applied in the present study, as differences in texture (not measured here) and geological parent material are more important for variations of background (metal) concentrations in pre-modern soils (Elberling et al. 2010). Indeed, similar clusters of elemental enrichment (P, Cu and Pb) are common in archaeological urban sites where artisanal dump and domestic waste have been recorded (e.g., McConnell et al. 2018, Holdridge et al. 2021). Soil micromorphological analysis characterised four main types of materials and associated activities, which reflect differences in use of space but also in the pace at which activities and processes took place in both the guild house and the main street. These findings are consistent with the timing of establishment and use determined by the Bayesian modelling. GUILD HOUSE The new data shows that the use of space changed over time and across different areas in the guild house (Fig. 9). As mentioned, excavation detected the presence of graves preceding the establishment of the guild house, suggesting the plot was part of a churchyard. The micromorphology record of a cess or latrine waste (presumably animal excrement), which had gradually built up in situ, suggests that the area was unoccupied by buildings, and perhaps served as a waste management or dump area before the establishment of the house. In this scenario, the plot was first used as a churchyard that was later abandoned, and not prepared for erecting the guild house. In this respect, geochemical results show elemental enrichment in all layers of the guild house, especially phosphorus (P) and lead (Pb), suggesting that the house was constructed in an area that had already been contaminated. This sequence shows that the use of plots even in the centre of a fully developed town could be more inconstant than what is indicated by archaeological excavation records and historical sources. Micromorphological evidence from the earliest occupation deposits show extensive cooking waste and debris, consistent with excavation records of hearths and phosphorus (P) enrichment detected in these levels, but also the presence of latrine waste in the same levels. This finding might be reflecting ground levelling as suggested by excavation records. The main occupation sequence of the guild house is characterised in thin section by a series of (micro-)floors and floor maintenance indicators, and evidence of cooking activities, domestic waste management and (indoor) traffic. Rather than a gradual floor build-up, as originally interpreted at excavation stage, the micromorphological records show that these deposits (BB), charcoal (CH), ash and fragments of cess, likely originating from rapid accumulation and dumping, OBM 19.3a. Centre: eastern area, OBM 1, thin section scan; c. and d. high phosphatic material rich in plant-derived organic fragments in a sub-unit composed of latrine (sub-unit a). Bottom: Western area, entrance hall OBM 20, monolith and thin section scan; e. mortar floor; f. mixed waste rich in high phosphatic material, wood charcoal, and ash. Microphotograph scale bars: 1 mm. Developer-Led Excavations 19 Fig. 7. Micromorphology of the main street. Top: OMB 23, monolith sample and thin section scan; a. peat-like material consisting of anthropic waste and plant residues, showing compression (arrows), and b. rare articulate phytoliths (elongate dendritic), OMB 23.1. Bottom: OMB 10, thin section scan; c. anthropic waste with common plant remains, excremental matter, burnt (fish) bone (BB), wood charcoal (CH) and d. rarely seed remains (S). Microphotograph scale bars: 1 mm. are the result of floor making and maintenance throughout occupation. To establish a definite count of floors observed in thin section is problematic because of their minute thickness (often less than 0.5 cm), micro-stratigraphic and spatial distribution (the same floor material appears in thin sections from different deposits, and not all can be securely linked to one level across different spaces). The micromorphological results, nonetheless, show the presence of three main types of floors (mortar; calcareous; beaten earth) that recur in multiple, distinctive levels. Some of these floors might be associated to a period prior or in between times when the building was used as a guild house, others are most likely contemporary to such a use. As mentioned, the guild house served as a place for gatherings for members of the guild on special occasions. Even though we cannot date the pace of floor maintenance, one enticing hypothesis is that floors were renewed prior to the festivities related to holidays. This would even take place in the secondary areas of the house, where food and drink for the festivities were prepared. The new modelled dating of the guild house predates its establishment to before 1435, when the guild was created. Perhaps the foundation year, 1435, given in the written sources, only marks a change in owner of the house, but not its use. 20 Federica Sulas et al. THE MAIN STREET In the main street area, organic-rich, anthropic waste was found mixed in with peat-like material, which appeared relatively well preserved due to postdepositional waterlogging. There was also indication of cess waste in this deposit. However, it remains unclear whether this can be associated with the presence of a (?) dunghill, simply derived from more sporadic dumping of waste or was intentionally added to absorb moisture. Noteworthy, Fig. 8. Bayesian statistical models on radiocarbon dates from the guild house (top) and the main street (bottom). ‘Boundary’ indicates the earliest events, the transition between phases and the latest events. The posterior probability distribution is marked in dark grey and the unmodelled calibrated dates are shown in light grey. The agreement index (A-value) indicates the match between the data and the model. For individual samples, see Supplementary Table SA. Developer-Led Excavations this deposit also included nutlets and fish bones (excreted as waste). Elsewhere in Odense, evidence of pig excrements, harbour porpoise and hop used for beer have been found in late 15th century-latrine deposits (Søe and Hoon Shin 2018). Enriched chemical levels (Cu, Sn, Pb and P) in one sample (X11516) from the deposit beneath the street might originate from domestic waste, as no evidence of a metal workshop was observed in the excavation here. Findings in Viking Age settlements in Norway (Cannell et al. 2020) and Denmark (Trant et al. 2020, 2021, 2024) suggest that identification of non-ferrous metalworking is possible by soil geochemistry in even more truncated contexts. Whilst the new results presented here cannot determine the ‘why’ of domestic waste in the main street area, they show that such waste did not accumulate in situ but was, in fact, dumped here. A sporadic, purposeless dumping would seem improvident as the first rain would have washed it away, leaving a trail of slippery mud. A purposeful waste dump would soon turn into an environmental hazard if left exposed and unattended – a scenario in contrast to street cleaning and waste management practices attested in medieval Scandinavian and other towns (see e.g., Jørgensen 2008). The organic-rich waste, identified in street thin sections, could have acted like a pad absorbing moisture and filling in space between the underlying material and the street level. The new results indicate that the main street was initially laid out. Around 1100, the level of the street was raised by dumping of domestic waste – rather than using gravel from outside the town as indicated in the paved street. Using waste material would require collection of the waste and, thus, a more nuanced view on waste as not 21 merely a problem but also a resource in the medieval city. PERSPECTIVES FOR INTEGRATING GEOARCHAEOLOGY IN DEVELOPER-LED EXCAVATIONS The results of this study show the great potential of applying geoarchaeological methods and analyses in developer-led excavations in Denmark and, indeed, beyond. Whilst the new geoarchaeological evidence clarifies formation and post-depositional processes in medieval Odense, it also raises new questions about urban space and waste in early towns. As the range of scientific methods boom, developer-led excavation budgets do not always follow the same rate. In Denmark, radiocarbon dating, dendrochronology and other well-established methods often prevail in bidding budgets, while geoarchaeological approaches are still considered new. This situation constrains development of competence, best practices and comparable analysis. Another challenge is connecting the micro-scale analysis of geoarchaeology with the macro-scale interpretations of field archaeology. It can be argued that micromorphology only shows what happened on the exact spot and is difficult to transfer it to a bigger scale. However, using the field archaeological interpretation of layers as a sort of ‘translator’, it is possible to upscale and transfer the interpretations on a microscale to similar layers in other areas of the excavated area. Micromorphology demands great care and minute documentation during fieldwork or else correlating field and laboratory records, and ultimately qualitative interpretation can be compromised. This sounds straightforward in principle but in our experience, it is more difficult to handle in practice. 22 Federica Sulas et al. Fig. 9. Schematic model of the medieval footprint at I. Vilhelm Werners plads, Odense. CONCLUSIONS: MEDIEVAL ODENSE THROUGH THE LENS OF A GUILD HOUSE AND A STREET Characterisation of archaeological sediments and microstratigraphy via soil geochemistry and micromorphology, and Bayesian modelling of radiocarbon dates elucidate the nature, space and pace of medieval occupation in central Odense. The new records reveal details of changing spatial organisation and waste management throughout the establishment, use and abandonment of a guild house along the main street in Odense. Evidence of cess and latrine waste dating to prior to the establishment of the guild house contributes new evidence to understanding the changing nature and functions of open areas in the town: from cemetery to wasteland and then guild house. Following the establishment of the guild house, microstratigraphic records capture a diversity of materials and processes that can be linked to selective uses and practices in the different parts of the building. Evidence of cooking activities, domestic waste management, and traffic inside the guild house magnify and expand the resolution of excavation records suggesting these. In particular, the microscale records reveal a sequence of floor making and maintenance, and waste management throughout the occupation of the guild house, rather than a gradual floor build-up as originally interpreted at excavation stage. On the main street, the new records question previous interpretations based on excavation findings of brick rubble and domestic waste. The micromorphological evidence shows that this waste was dumped and experienced prolonged periods of waterlogging conditions, enabling the preservation of organic remains. The dumping of domestic waste here might be reflecting a change of urban maintenance practises about 50 years after the establishment of the paved main street. The geoarchaeological methods employed in this study provide answers to some of the questions raised by the field excavations and reveal new evidence to reconsider initial interpretations. Prior to geoarchaeological analyses, Developer-Led Excavations interpretations associated the excavation deposits to a guild house and a paved street, dating to different periods. Geoarchaeological analyses and Bayesian work, presented here, define conditions in place before the establishment of these structures, and use and tempo of space and maintenance practices previously undocumented. A challenge that one must consider very carefully, however, is how to link the macro-spectrum of the excavation (layers, units) with the micro-spectrum of soil micromorphology and geochemistry. Close cooperation between archaeologists and geoarchaeologists is key to establish which hypotheses the analyses are likely to confirm or refute, and how the scientific results should be interpreted. For field archaeology to fully profit from geoarchaeological methods, urban excavations need to integrate them from the start. This could be an economic challenge in the ways developer-led excavations are financed in Denmark, but the prospects of the outcome are promising. ACKNOWLEDGEMENTS We extend our gratitude to: the OBM9776 excavation team for exemplar sampling that made our analyses possible; the Centre for Urban Network Evolutions (UrbNet), Aarhus University and its leaders R. Raja and S.M. Sindbæk for support; Thomas Ljungberg for guiding and helping with the geostatistical analyses; Sara Pesci (Quaternaria) for providing first descriptions of some of the thin sections analysed; the C. McBurney Laboratory for Geoarchaeology, University of Cambridge, for thin section processing. Last but not least, we are grateful for detailed and insightful comments provided by two external reviewers, and the journal editor, which have sharpened our argument and improved the paper. DISCLOSURE STATEMENT No potential conflict of interest was reported by the authors. 23 DATA AVAILABILITY STATEMENT Data that support the findings of this study are openly available at the CENTRUM Forskningscenter for centralitet, Odense Bys Museer: Rapport n. 10: http://download.obm.dk/ files/publikationer/Rapport-10-Soil% 20micromorpholgy%20in%20medieval% 20Odense.pdf Rapport n. 11: http://download.obm.dk/ files/publikationer/Rapport-11-Soil%20geo chemistry%20in%20medieval%20Odense.pdf SUPPLEMENTARY MATERIAL Supplemental data for this article can be accessed online at https://doi.org/10.1080/ 00293652.2025.2515515. ORCID Federica Sulas http://orcid.org/0000-00031825-8133 Cristiano Nicosia http://orcid.org/0000-00024716-8893 Kirstine Haase http://orcid.org/0000-00024288-3169 Mikael Manøe Bjerregaard http://orcid.org/ 0000-0002-5352-6204 Søren Munch Kristiansen http://orcid.org/ 0000-0003-3128-4061 REFERENCES Angelucci, D., 2017. Lithic artefacts. In: C. Nicosia and G. Stoops, eds. Archaeological soil and sediment micromorphology. Chichester, UK: Wiley, 223–230. Banerjea, R.Y., et al. 2016. A multi-proxy, diachronic and spatial perspective on the urban activities within an indigenous community in medieval Riga, Latviab. Quaternary International, 460, 3– 21. doi:10.1016/j.quaint.2017.02.006 Bjerregaard, M.M., 2019. OBM9776 I. Vilhelm Werners Plads, Odense sogn, Odense herred, tidl. Odense amt. Sted nr. 08.04.07. Sb.nr. 130. Kampagne: 13-01-2016. Available from: https:// www.kulturarv.dk/publicffdata/documenta tion/file/doc/215869/public 24 Federica Sulas et al. Brink, S., 1996. Political and social structures in early Scandinavia. A settlement-historical pre-study of the central place. Tor, 28, 235–281. Bronk Ramsey, C., 2009. Bayesian analysis of radiocarbon dates. Radiocarbon, 51 (1), 337–360. doi:10.1017/S0033822200033865 Bullock, P., et al. 1985. Handbook for soil thin section description. Albrighton, UK: Waine Research. Cammas, C., 2018. Micromorphology of earth building materials: toward the reconstruction of former technological processes (protohistoric and historic periods). Quaternary International, 483, 160–179. doi:10.1016/j. quaint.2018.01.031 Cannell, R.J.S., et al. 2020. Geochemical analysis of the truncated Viking age trading settlement of Heimdalsjordet, Norway. Geoarchaeology, 35 (5), 748–771. doi:10.1002/gea.21795 Canti, M.G., 2017a. Charred plant remains. In: C. Nicosia and G. Stoops, eds. Archaeological soil and sediment micromorphology. Chichester, UK: Wiley, 141–142. Canti, M.G., 2017b. Burnt carbonates. In: C. Nicosia and G. Stoops, eds. Archaeological soil and sediment micromorphology. Chichester, UK: Wiley, 181–188. Christensen, A.S., 1988. Middelalderbyen Odense. Copenhagen: Centrum. Christensen, J.T., et al. 2019. Odense before and after the canonization of Cnut. In: S. Hope, ed. Life and cult of Cnut the holy - The first royal saint of Denmark. Report from an interdisciplinary seminar in Odense, November 6th to 7th 2017. Odense: University Press of Southern Denmark and Odense City Museums, 10–25. Christophersen, A., 2015. Performing towns. Steps towards an understanding of medieval urban communities as social practice. Archaeological Dialogues, 22 (2), 109–132. doi:10.1017/S1380203815000161 Council of Europe, 1992. European convention on the protection of the archaeological heritage (revised). European treaty series No. 143. Available from: https://rm.coe.int/168007bd25 [Accessed 29 March 2025]. Daly, A., 2019. Dateringsundersøgelse af tømmer fra Thomas B. Thriges Gade, Odense. Dendro. dk rapport 47. Dejmal, M., and Brown, T., 2014. Medieval horse stable; the results of multi proxy interdisciplinary research. PLOS ONE, 9 (3), e89273. doi:10. 1371/journal.pone.0089273 Devos, Y., et al. 2013. Studying urban stratigraphy: dark earth and a microstratified sequence on the site of the court of Hoogstraeten (Brussels, Belgium). Integrating archaeopedology and phytolith analysis. Quaternary International, 315, 147–166. doi:10.1016/j. quaint.2013.07.024 Devos, Y., Nicosia, C., and Wouters, B., 2020. Urban geoarchaeology in Belgium: experiences and innovations. Geoarchaeology, 35 (1), 27–41. doi:10.1002/gea.21755 Elberling, B., et al. 2010. Heavy metals in 3300-year-old agricultural soils used to assess present soil contamination. European Journal of Soil Science, 61 (1), 74–83. doi:10.1111/j. 1365-2389.2009.01202.x Engelmark, R., and Linderholm, J., 2008. Miljöarkeologi: Människa och landskap – ett komplicerad dynamic. Malmö: Malmö Kulturmiljö. French, C., and Rajkovaca, T., 2015. A step-bystep guide to the making of soil thin sections. McBurney Geoarchaeology Laboratory, University of Cambridge. https://www.arch. cam.ac.uk/files/guide-thin-sections.pdf Gebhardt, A., and Langohr, R., 1999. Micromorphological study of construction materials and living floors in the Medieval Motte of Werken (West Flanders, Belgium). Geoarchaeology, 14 (7), 595–620. doi:10.1002/ (SICI)1520-6548(199910)14:7<595::AIDGEA1>3.0.CO;2-Q Goslar, T., 2015. Description of procedures of AMS 14C dating used in the poznań radiocarbon laboratory Available from: https://radiocarbon. pl/wp-content/uploads/2018/07/procedure_ ams_prl.doc) [Accessed 12 January 2022]. Haase, K., 2017. OBM9776 I. Vilhelm Werners Plads, Odense sogn, Odense herred, tidl. Odense amt. Sted nr. 08.04.07. Sb.nr. 130 - Kampagne: 16-05-2013. Available from: https://www.kultur arv.dk/publicffdata/documentation/file/doc/ 185976/public Haase, K., 2019a. An urban way of life. Social practices, networks and identities in Odense, AD 1000–1500. Unpublished thesis. Aarhus University. Haase, K., 2019b. Building urbanity: spatial organisation as social practice in medieval Odense. Acta Archaeologica, 90 (2), 43–75. doi:10.1111/ j.1600-0390.2019.12212.x Developer-Led Excavations Haase, K., 2022. Animal husbandry, import replacement, and urban growth in medieval Odense, Denmark. Journal of Urban Archaeology, 5, 211–227. doi:10.1484/J.JUA.5. 129850 Haase, K., and Olsen, J., 2021. Building a high-resolution chronology of a medieval urban site through Bayesian modelling. Archaeometry, 63 (4), 860–877. doi:10.1111/ arcm.12651 Holdridge, G., et al., 2021. A Roman provincial city and its contamination legacy from artisanal and daily-life activities. PLOS ONE, 16 (6), e0251923. doi:10.1371/journal.pone.0251923 Jørgensen, D., 2008. Cooperative sanitation: managing streets and gutters in late Medieval England and Scandinavia. Technology and Culture, 49 (3), 547–567. doi:10.1353/tech.0. 0047 Kenward, H., and Hall, A., 2008. Urban organic archaeology: an irreplaceable palaeoecological archive at risk. World Archaeology, 40 (4), 584–596. doi:10.1080/00438240802453286 Kristiansen, S.M., and Dalsgaard, K., 2000. Soil evolution in the remnants of natural forest vegetation: an example from an old oak-lime coppice forest in Denmark. Geografisk Tidsskrift-Danish Journal of Geography, 100 (1), 27–36. doi:10.1080/00167223.2000. 10649436 Kristiansen, S.M., Ljungberg, T.L., and Sulas, F., 2021. Soil geochemistry in medieval Odense analysis of samples from the Guild House and the Old Road area (OBM 9776 II). CENTRUM forskningscenter for centralitet, rapport nr. 11. Odense, Odense Bys Museer. Linderholm, J., et al. 2019. Ørlandet iron age settlement pattern development: geoarchaeology (geochemistry and soil micromorphology) and plant macrofossils. In: I. Ystgaard, ed. Environment and settlement: Ørland 600 BC– AD 1250. Oslo: Cappelen Damm, 107–134. Linderholm, J., 2021. The Sørenga D1A borehole site, Oslo harbour, Norway: a multi-analytical geoarchaeological and palaeoenvironmental approach. In: C. Bennet and T. Walker, eds. Environment, archaeology and landscape: papers in honour of professor Martin Bell. Oxford: Archaeopress, 65–76. Linderholm, J., et al. 2023. Soil chemical analysis on samples from a medieval moat at Paradis 51. Interim report. Umeå: Umeå universitet. 25 Miljöarkeologiska Laboratoriet Rapport nr. 2023-007. Lund, Scania, Sweden. Macphail, R.I., et al. 2013. Integrated microstratigraphic investigations of coastal archaeological soils and sediments in Norway; the gokstad ship burial mound and its environs including the viking harbour settlement of Heimdaljordet, Vestfold. Quaternary International, 315, 131–146. doi:10.1016/j.quaint.2013.05.051 Macphail, R.I., et al. 2016. E6 gudbrandsdalen project (Brandrud, Fryasletta, Grytting, and Øybrekka), oppland, Norway: soil micromorphology (with selected microchemistry, bulk soil-chemistry, carbon-polymer, particle-size, and pollen analyses. In: I.M. Gundersen, ed. Gård og utmark i Gudbrandsdalen: Arkeologiske undersøkelser i Fron 2011–2012. Oslo: Kulturhistorisk museum, Arkeologisk seksjon, 304–317. Macphail, R.I., and Goldberg, P., 2018. Applied soils and micromorphology in archaeology. Cambridge: Cambridge University. Macphail, R.I., Linderholm, J., and Eriksson, S., 2015. Odense, Funen, Denmark (OBM 9776: I. Vilhelm Werners Plads): soil micromorphology, chemistry and magnetic susceptibility. Odense, Denmark, Report for Odense Bys Museer. Macphail, R.I., Lindholm, J., and Gjerpe, L.E., 2022. Speculations on farming development during the early iron age of southern Norway (500 BC–AD 550), focusing on the dobbeltspor dilling project. In: F. Sulas, H. Lewis and M. Arroyo-Kalin et al. eds. Inspired geoarchaeologies: past landscapes and social changes. Cambridge: McDonald Institute for Archaeological Research, 145–155. Matthews, W., et al. 1997. Microstratigraphic traces of site formation processes and human activities. World Archaeology, 29 (2), 281–308. doi:10.1080/00438243.1997.9980378 Mazurek, R., et al. 2016. Micromorphological and physico-chemical analyses of cultural layers in the urban soil of a medieval city — a case study from Krakow, Poland. Catena, 141, 73–84. doi:10.1016/j.catena.2016.02.026 McConnell, J., et al. 2018. Lead pollution recorded in Greenland ice indicates European emissions tracked plagues, wars, and imperial expansion during antiquity. Proceedings of the National Academy of Sciences, 115 (22), 5726– 5731. https://doi.org/10.1073/pnas.1721818115. 26 Federica Sulas et al. Milek, K.B., 1997. Soil micromorphology and the medieval urban environment: examples from Ely and Peterborough, England. In: G. DeBoe and V. Verhaeghe, eds. Environment and subsistence in Medieval Europe: papers of the ‘Medieval Europe brugge 1997 conference. Vol. 9. Zellik, Belgium: Institute of Archaeological Heritage, pp. 155–168. Milek, K.B., 2012. Floor formation processes and the interpretation of site activity areas: an ethnoarchaeological study of turf buildings at Thverá, Northeast Iceland. Journal of Anthropological Archaeology, 31 (2), 119–137. doi:10.1016/j.jaa.2011.11.001 Miljøstyrelsen, 2018. Liste over kvalitetskriterier i relation til forurenet jord. Copenhagen: By- og Landskabsstyrelsen. https://mst.dk/erhverv Nicosia, C., et al. 2012. Medieval dark earth in an active alluvial setting from the Uffizi Gallery complex in Florence, Italy. Geoarchaeology, 27 (2), 105–122. doi:10.1002/gea.21403 Nicosia, C., and Stoops, G., eds. 2017. Archaeological soil and sediment micromorphology. Chichester, UK: Wiley. Petersén, A.H., Utigard Sandvik, P., and Sveistrupc, T.E., 2015. The rural–urban connection as revealed in the terrestrial sediments at Nidarneset, Trondheim, Norway. Journal of Archaeological Science: Reports, 2, 389–404. doi:10.1016/j.jasrep.2015.03.008 Reid, V., et al. 2023. The role of geoarchaeology in the interpretation of fragmented buildings and occupation surfaces: the case of coastal settlements in northeast Scotland. Geoarchaeology, 39 (3), 238–267. doi:10.1002/gea.21990 Reimann, C., et al. 2014. Chemistry of Europe’s agricultural soils, part a methodology and interpretation of the GEMAS data set. Geologische Jahrbuch, Reihe B, 102. Stuttgart, Germany: E. Schweizerbart’sche https://www. schweizerbart.de/publications/detail/isbn/ 9783510968466 Reimer, P., et al. 2020. The IntCal20 northern hemisphere radiocarbon age calibration curve (0–55 cal kBP). Radiocarbon, 62 (4), 725–757. doi:10.1017/RDC.2020.41 Rentzel, P., 2017. Trampling, poaching and the effect of traffic. In: C. Nicosia and G. Stoops, eds. Archaeological soil and sediment micromorphology. Chichester, UK: Wiley, 281–295. Runge, M., and Henriksen, M., 2018. The origins of Odense - new aspects of early urbanisation in southern Scandinavia. Danish Journal of Archaeology, 7 (1), 1–67. doi:10.1080/ 21662282.2018.1475891 Sageidet, B.M., 2000. Soil micromorphology and its contribution to the interpretation of archaeological sites. In: L. Selsing, ed. Norsk kvartærbotanikk ved årtusenskiftet. AmSVaria. Vol. 37. Stavanger: Museum of Archaeology, 21–25. Sageidet, B.M., 2013. Soil micromorphological evidence from iron age land use at Tornby and Mörtlösa in Linköping, Sweden. Quaternary International, 315, 116–130. doi:10.1016/j.quaint.2013.07.012 Shahack-Gross, R., et al. 2005. Geoarchaeology in an urban context: the uses of space in a Phoenician monumental building at Tel Dor (Israel). Journal of Archaeological Science, 32 (9), 1417–1431. doi:10.1016/j.jas.2005.04.001 Shelley, A., 2005. Dragon hall, King street, Norwich: excavation and survey of a late Medieval Merchant’s trading complex. East Anglian archaeology report No. 112. Norwich, Norfolk and Norwich Heritage Trust. Søe, M.J., and Hoon Shin, D., 2018. Ancient DNA from latrines in Northern Europe and the Middle East (500 BC–1700 AD) reveals past parasites and diet. PLOS ONE, 13 (4), e0195481. doi:10.1371/journal.pone.0195481 Stoops, G., 2003. Guidelines for analysis and description of soil and regolith thin sections. Madison, WI: Soil Science Society of America. Stoops, G., Canti, M., and Kapur, S., 2017. Calcareous mortars, plasters and floors. In: C. Nicosia and G. Stoops, eds. Archaeological soil and sediment micromorphology. Chichester, UK: Wiley, 189–199. Stoops, G., Marcelino, V., and Mees, F., eds. 2010. Interpretation of micromorphological features of soils and regoliths. Amsterdam: Springer. Sulas, F., et al. 2021. Soil micromorphology in medieval Odense - analysis of samples from the Guild House and the Old Road area (OBM 9776 II). CENTRUM forskningscenter for centralitet, rapport nr. 10. Odense, Odense Bys Museer. Sulas, F., et al. 2022. Revealing the invisible dead: integrated bio-geoarchaeological profiling exposes human and animal remains in a seemingly ‘empty’ viking-age burial. Journal of Archaeological Science, 141, 105589. doi:10. 1016/j.jas.2022.105589 Developer-Led Excavations Trant, P.L.K., et al. 2021. How sampling resolution affects the outcome: a geochemical study of a Viking age house in Ribe, Denmark. Archaeological and Anthropological Science, 13 (1), 21. doi:10.1007/s12520-020-01243-7 Trant, P.L.K., et al. 2024. A multi-proxy geochemical and micromorphological study of the use of space and stratigraphy of a Viking-age house in Ribe, Denmark. Archaeological and Anthropological Science, 16 (4), 59. doi:10.1007/s12520-024-01962-1 27 Trant, P.L.K., Kristiansen, S.M., and Sindbæk, S. M., 2020. Visible near-infrared spectroscopy as an aid for archaeological interpretation. Archaeological and Anthropological Science, 12 (12), 280. doi:10.1007/s12520-020-01239-3 Wouters, B., et al. 2017. Medieval markets: a soil micromorphological and archaeobotanical study of the urban stratigraphy of Lier (Belgium). Quaternary International, 460, 48–64. doi:10.1016/j.quaint.2017.03.002