SHORT REPORT Sem: A Viking Age metalworking site in the southeast of Norway? Lars Gustavsen 1 | Monica Kristiansen 1 | Erich Nau 1 | Bernt Egil Tafjord 2 1 The Norwegian Institute for Cultural Heritage Research, Oslo, Norway 2 Øvre Eiker Municipality, Hokksund, Norway Correspondence Lars Gustavsen, The Norwegian Institute for Cultural Heritage Research, Storgata 2, 0105 Oslo, Norway. Email: lars.gustavsen@niku.no Funding information Buskerud County Municipality Abstract Evidence pertaining to Viking Age industry and trade is rarely encountered in the Norwegian archaeological record, although recent evidence suggests that sites of this character are more common than previously thought. Systematic metal detecting and large‐scale geophysical surveys carried out in Øvre Eiker in the southeast of Norway between 2014 and 2016 revealed a multiphase and complex settlement area, consisting of large pits potentially associated with metalworking. The evidence so far points to a site with far‐reaching contemporary connections, where objects were worked and exchanged. This short report aims to present the results from the work undertaken so far and to demonstrate how the complementary use of geophysical surveys and metal detection can add cultural‐historical depth to the site itself and to a little understood aspect of Early Medieval history. KEYWORDS early medieval, ground‐penetrating radar, metal detecting, metalworking, Norway, Viking age 1 | BACKGROUND The farm Sem is situated some 2.5 km southwest of the town of Hokksund in Øvre Eiker Municipality, southeast Norway. It sits atop a low, northeast‐running and gently sloping ridge in a fertile landscape at the base of the glacial Drammen Valley. The valley itself contains the Drammen River, which is again connected to the sea by way of the Drammen Fjord and the Oslo Fjord (Figure 1). Although the Eiker district is generally rich in archaeological finds and scheduled monuments from all time periods, it is perhaps best known for the spectacular late ninth century Hoen hoard, uncovered in 1834 some 3 km north of Sem. One of the largest Viking hoards ever found, it comprises a range of gold and silver‐gilt objects, as well as 20 coins of diverse provenance, all secondarily reworked into pendants (Fuglesang & Wilson, 2006). In the immediate vicinity of the study area surviving Iron Age burial mounds are documented southwest of the site, as is a defunct medieval church site. Sundry stray finds have been made by metal detectorists in the surrounding fields, of which an Arabic coin of the late eighth century and a copper alloy fitting imitating Carolingian plant ornamentation are of interest. These finds gave intriguing hints as to the importance of Sem in the Early Medieval period, but it was not until 2014, when an unusual finds assemblage was recovered by metal detecting, that the true significance of the site was realized. The assem- blage comprised scale weights, Arabic coins, hack silver, and fragments of brooches and book fittings, as well as production waste in the form of slags and crucible fragments, pointing to a site involved in trade and pro- duction (Figure 2). In light of these discoveries a ground‐penetrating radar (GPR) survey was carried out in 2015 with the aim of investigating the sub‐surface deposits from which the finds originated. Based on the results from this survey, a systematic metal detecting campaign was then mounted in 2016 to refine the dating and function of the site, resulting in further finds of similar character to those previously found. The aim of this short report is to present the preliminary results from the project, and to compare these with evidence from previously excavated sites. Thus, we will forward a tentative interpretation of some of the fea- tures found by geophysical methods, as well as of the site itself. Further- more, the report will demonstrate how the complementary use of systematic metal detecting and geophysical methods can be used to gain further knowledge of a site without resorting to intrusive methods. [Correction added on 19 October 2018, after first online publication: the first paragraph of the Methods section has been updated in this version]. Revised: 3 August 2018 Accepted: 12 September 2018 DOI: 10.1002/arp.1726 Archaeological Prospection. 2019;26:13–20. © 2018 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/arp 13