EAC occasional paper no. 4 no. paper occasional EAC

EAC Occasional Paper No. 5 EAC occasional paper no. 4 EAC Occasional Paper No. 5 Remote Sensing for Archaeological HeritageHeritage ManagementManagement of Farmed HeritageRemote SensingManagement for and Forested Landscapes in Europe of Farmed and Forested Edited by David C Cowley Europe in Landscapes Forested and Farmed of Management Heritage Archaeological Heritage Edited by Stephen Trow, Vincent Holyoak and Emmet Byrnes LandscapesManagement in Europe

RemoteSome 40 sensing per cent is of oneEurope of is the farmed main and foundations 47 per cent offorested. archaeological The future of forRemote Sensing Archaeological Heritage Management data, under pinning knowledge and understanding of the historic the majority of Europe’s archaeological sites therefore depends on rural land environment. The volume, arising from a symposium organised by the Europaeuses that Archaeologiae lie outside the spatial Consilium planning (EAC) and and development the Aerial control Archaeology systems of Researchits various Group nation (AARG), states. Thisprovides volume, up producedto date expert by the statementsEuropean Association on the of methodologies,Archaeologists (EAA) achievements and Europae and Archaeologiae potential of remoteConsilium sensing (EAC) Jointwith Working a particular focus on archaeological heritage management. Well-established Group on Farming, Forestry and Rural Land Management, examines the challenges approaches and techniques are set alongside new technologies and data-sources,posed by agriculture, with discussion forestry andcovering other relativerural land merits uses inand terms applicability, of the long-term andconservation the need forof Europe’sintegrated archaeological approaches sitesto understanding and the management and managing of its historic thelandscapes. landscape. Discussions cover aerial photography, both modern and historic, LiDAR, satellite imagery, multi-and hyper-spectral data, sonar and geophysical survey, addressing both terrestrial and maritime contexts. Case studies drawn from the contrasting landscapes of Europe illustrateEAC Occasional best practice Paper andNo. 4innovative projects. ISBN 978-963-9911-17-8

EAC Occasional Paper No. 5 Occasional Publication of the Aerial Archaeology Research Group No. 3 Vincent Holyoak and Emmet Byrnes Emmet and Holyoak Vincent Trow,Stephen by Edited ISBN 978-963-9911-20-8 Edited by David C Cowley

Edited by Stephen Trow, Vincent Holyoak and Emmet Byrnes Edited by David C Cowley

EAC5_cover.indd 1 2011.02.22. 16:53:53 EAC Occasional Paper No. 5 Occasional Publication of the Aerial Archaeology Research Group No. 3 Remote Sensing for Archaeological Heritage Management EAC Occasional Paper No. 5

Occasional Publication of the Aerial Archaeology Research Group No. 3

Remote Sensing for Archaeological Heritage Management

Proceedings of the 11th EAC Heritage Management Symposium, Reykjavík, , 25­27 March 2010

Edited by David C Cowley EAC Occasional Paper No. 5 Occasional Publication of the Aerial Archaeology Research Group No. 3 Remote Sensing for Archaeological Heritage Management Edited by David C Cowley

Published by: Europae Archaeologia Consilium (EAC), Association Internationale sans But Lucratif (AISBL), Siège social Koning Albert II-laan '" Avenue Roi Albert II '" P.O. Box ') Boîte ') '(') Brussel '(') Bruxelles Belgium Belgique www.e-a-c.org

In association with: Aerial Archaeology Research Group

© The individual authors ()''

The opinions expressed in this volume are those of the individual authors, and do not necessarily represent o*cial policy.

ISBN "#$-"%&-""''-()-$

Brought to publication by Archaeolingua, Hungary Managing editor: Elizabeth Jerem Copy editing by Dorottya Domanovszky Layout and cover design by Gergely Hős Printed by Aduprint Printing and Publishing Ltd, Hungary Distribution by Archaeolingua, Hungary

Cover image: Airborne Laser Scan (LiDAR) of a forested area before and after !ltering (St. Anna in der Wüste, Austria). © Michael Doneus and Klaus Löcker, LBI-ARCHPRO, Vienna Contents

Foreword # Katalin Wollák, President of Europae Archaeologiae Consilium

Acknowledgments $ David C Cowley

Opening address " Katrín Jakobsdóttir, Minister of Education, Science and Culture, Iceland

' | Remote sensing for archaeological heritage management '' David C Cowley and Kristín Huld Sigurðardóttir

Making remote sensing work for archaeological heritage management

( | Identifying the unimaginable – Managing the unmanageable '# Dominic Powlesland

& | ‘Total Archaeology’ to reduce the need for Rescue Archaeology: The BREBEMI Project (Italy) && Stefano Campana

, | Remote sensing for archaeology and heritage management – site discovery, ,& interpretation and registration David C Cowley

New environments and technologies: challenges and potential

- | Airborne Laser Scanning in forested areas – potential and limitations of -" an archaeological prospection technique Michael Doneus and Christian Briese

% | High resolution LiDAR speci!cally for archaeology: are we fully exploiting this valuable resource? ## Robert Shaw and Anthony Corns

# | Archaeological applications of multi/hyper-spectral data – challenges and potential $# Anthony Beck

$ | Making the most of airborne remote sensing techniques for archaeological survey and interpretation "" Rebecca Bennett, Kate Welham, Ross A Hill and Andrew Ford

" | &D recording for cultural heritage ')# Fabio Remondino

') | Through an imperfect !lter: geophysical techniques and the management of archaeological heritage ''# Chris Ga"ney and Vincent Ga"ney

'' | Marine geophysics: integrated approaches to sensing the seabed '(" Antony Firth

Exploring the archaeological resource base

'( | The English Heritage National Mapping Programme ',& Pete Horne

'& | Integrating survey data – the Polish AZP and beyond '-& Włodek Rączkowski % EAC OCCASIONAL PAPER NO. -

', | As far as the laser can reach…: Laminar analysis of LiDAR detected structures as a powerful '%' instrument for archaeological heritage management in Baden-Württemberg, Germany Jörg Bo%nger and Ralf Hesse

'- | Between the Lines – enhancing methodologies for the exploration of extensive, '#& inundated palaeolandscapes Simon Fitch, Vincent Ga"ney, Benjamin Gearey and Eleanor Ramsey

'% | Aerial archives for archaeological heritage management: The Aerial Reconnaissance ()- Archives – a shared European resource Lesley Ferguson

Using remote sensed data: interpretation and understanding

'# | Remote sensing for the integrated study and management of sites and monuments ('- – a Central European perspective and Czech case study Martin Gojda

'$ | Airborne Laser Scanning for the management of archeological sites in Lorraine (France) ((" Murielle Georges-Leroy

'" | Aerial archaeological survey of a buried landscape: The Tóköz project (&- Zoltán Czajlik, László Rupnik, Máté Losonczi and Lőrinc Timár

() | The archaeological landscape of northeast Iceland: a ghost of a Viking Age society (,& Árni Einarsson and Oscar Aldred

(' | Reserved optimism: preventive archaeology and management of cultural heritage in Slovenia (-" Gašper Rutar and Matija Črešnar

(( | World War I Heritage in Belgium: combining historical aerial photography and EMI (%- Birger Stichelbaut, Timothy Saey, Fun Meeuws, Jean Bourgeois and Marc Van Meirvenne

(& | An aerial view of the past – aerial archaeology in Denmark (#- Lis Helles Olesen

(, | Knowledge-based aerial image interpretation ($& Rog Palmer

(- | Training and development: the next phase? ("& Chris Musson

Contributors (""

Résumés Catherine Fruchart &)'

Zusammenfassungen Johanna Dreßler &)# 20 | The archaeological landscape of northeast Iceland: a ghost of a Viking Age society

Árni Einarsson and Oscar Aldred

Abstract: A current mapping project of medieval archaeology in an area of over 3,000km2 in northeastern Iceland makes use of aerial photography on a large scale, including over 2,000 low altitude oblique photographs. Aided by ground surveys and other !eldwork, the project has revealed a whole system of well-preserved remains dating to the medieval period. An extensive network of turf walls is most prominent, but farmsteads, churches, assembly sites, pagan grave !elds, peat cutting, charcoal pits and horse tracks are also clearly visible. This medieval archaeological landscape belongs largely to the Viking period and is a relic of the settlement pattern of the !rst few generations in Iceland. Its high conservation value and future preservation is discussed in the paper.

Introduction

Iceland was discovered and colonised in the late "th oblique low altitude aerial photography supported century, following the westward expansion of the Norse by archaeological excavations and ground surveys. seafaring Viking culture. The settlement period and the Augmented by related projects, as well as existing subsequent three centuries have been immortalised knowledge (McGovern et al. #$$)), this e(ort gradually in the rich medieval saga literature of the . revealed a whole network of walls and other features This period was remarkable for the creation of a new (Figure #$.') that were roughly contemporaneous and society with a hierarchical system of assemblies with contained important elements of the Viking period legislative and judicial functions, and the distribution society. Apart from turf walls, these features included of power among a number of chieftains instead of one assembly sites, burial sites, churches/chapels and farm central authority. This political system was to last more sites of various characters, and also features that have than three centuries, until Iceland became part of the not been properly dated yet, like routes, peat cuttings Norwegian and later, Danish kingdom. and charcoal pits. Many of the oldest features are so well preserved, widespread and prominent that we Much of the early archaeological work in Iceland was have been tempted to describe this archaeological concentrated on the Viking period (Vésteinsson #$$%a; landscape as a relic of the Viking period. Friðriksson '""%) because it represents a ‘glorious past’ and also because the identity of the Icelandic nation is rooted in events and processes in this period. The Viking period is still a major focus in the archaeology of Iceland but, in line with global trends in science, the emphasis is now on the dynamic interaction of humans and their environment. This entails a greater focus on o(-site archaeology, primarily associated with the use of resources. This wider scope calls for a landscape approach with large scale dating and registration of monuments, preferably aided by remote sensing.

This paper provides the context for and the description of the !ndings from a research project that has used aerial photography on an unprecedented scale for Iceland. Initial research began as a study of an extensive system of medieval turf walls in northeastern Iceland, using vertical aerial photographs, revealing a complex division of the landscape (Einarsson et al. #$$#). The project then continued, involving wide ranging

Figure #$.': The great turf wall at Fljótsheiði. Note how the wall traces the edge of the bog to the right. The wall continues in the upper left corner of the photograph. Some of the main horse tracks can be seen as winding linear features and there are peat pits to the right. #%% EAC OCCASIONAL PAPER NO. *

In this paper we !rst give an overview of the use of use of aerial survey speci!cally for archaeology. aerial photography in Iceland for cartography and Sveinbjörn used both conventional and infra-red archaeology. We then focus on our study area in the images. However, the .ights were largely directed at northeast of Iceland, listing some of the most prominent previously known sites in order to obtain a basic record, types of ancient remains we see from the air, Viking and no detailed transcription mapping was involved. period or later, and !nally discussing issues relating to In the '",$s, Guðrún Sveinbjarnardóttir conducted heritage management and the conservation value and regional studies of farm abandonment in Iceland for preservation of the archaeological landscape. her doctoral thesis (Sveinbjarnardóttir '""#). She made occasional use of aerial photographs, although this was hampered by the high altitude of the photography The use of aerial observations in Icelandic and the di/culty in recognising distinct features. archaeology In the '""$s, the Institute of Archaeology, Iceland (Fornleifastofnun Íslands), amongst other professional The history of aerial photography in Iceland can archaeologists, began extensive !eld surveys. At be traced back to oblique photographs taken of !rst, aerial reconnaissance was incorporated into the Reykjavík in '"'". However, aerial photography was survey, taking oblique shots in both conventional and not systematically carried out until '"+)-,, when it infra-red format. This programme, however, was short- was used to compliment the ground surveys for the lived. !rst detailed maps of Iceland at a published scale of ':'$$,$$$ (Bragason & Guðmundsson '",,). World The !rst publication that presented the explicit use of War II marked an expansion in the areas covered by aerial photographs in archaeology in Iceland was in aerial photographs, again for cartography. However, '""* (Ísaksson & Helgason '""*). The article discussed these series of vertical aerial photographs taken by the the use of the technique in relation to features on farms German, British and American military, as well as later and land around Reykjavík and particularly highlighted by the National Land Survey of Iceland (Landmælingar the advantages of photography under light snow Íslands), are an invaluable resource in two related but cover. Photographs had been taken of the same site di(erent ways. The !rst is tracing landscape change under di(erent conditions, using both conventional over the last -$ years. This is a period that has seen and infra-red !lm. The article also attempted to discuss radical changes in the landscape, especially around the speci!c nature of aerial survey in Iceland as well as farms and urban centres. Secondly these aerial sources promote its more systematic use in archaeology. are invaluable in the mapping of archaeological sites. The retrospective mapping of archaeology from Although archaeological survey has a long tradition in the earliest available sources often provides a much Iceland, beginning in the '"th century (Friðriksson '""%), better understanding of landscape formation and it is only in the last quarter of the #$th century that the organisation and the status of archaeological features !rst systematic archaeological surveys of the country than using conventional map sources (Aldred et al. commenced. In the !rst decade of the #'st century #$'$). tremendous progress has been made and the number of sites registered increased dramatically. In the same In the late '")$s and early '",$s Sveinbjörn Rafnsson period, usage of aerial photographs in archaeological used oblique aerial photography as an integral part of survey multiplied and currently most archaeological his archaeological survey of deserted valleys in the east surveys use aerial photographs at some stage. In most of Iceland (Rafnsson '""$). This was the !rst practical cases, these are vertical aerial photographs, often from

Figure #$.#: A medieval turf wall. #$ The archaeological landscape of northeast Iceland: a ghost of a Viking Age society #%* a fairly high altitude of ',,$$$ft or *,%,-m, with a focal The aerial archaeology of northeast Iceland length of '*#mm (a scale of ':+-,$$$). The main project that encapsulates the foundation Recent survey, driven by new legislation in '",", has work and best practices for aerial survey is focused on aimed to achieve total coverage of districts and regions northeast Iceland. It was carried out by the Institute (Ólafsson '""'; Friðriksson & Vésteinsson '"",). The of Archaeology and the Mývatn Research Station and underlying premise of contemporary practice is to funded by the Icelandic Centre for Research (RANNÍS) systematically register all surviving remains to a set and Þjóðhátíðarsjóður (a fund that commemorates standard. The registration is informed by textual and the "th century ) to map the oral histories. Although these are invaluable sources extent and preservation of the ancient wall systems in in discerning the presence of archaeology, textual northeast Iceland. The aim was to determine their date, sources can be fragmentary in coverage, and the use of spatial extent, structure and possible function. The oral sources, derived from place-name surveys, is based project systematically employed aerial photographs, on an assumption of continuity in the social memory both verticals and obliques, as well as satellite imagery, from one generation to the next. Aerial photography to map the archaeology of a large area. At last count provides valuable additional records of sites which may over +,$$$km# have been covered and about %$$km of have been identi!ed by other means. It reveals certain walls have been mapped. types of sites (e.g. boundaries and charcoal pits) that tend to be under-represented in the other types of The study area evidence and which are di/cult to !nd or comprehend Our study area covers +,'-%km# in the Counties of Suður- through conventional !eld walking (Figure #$.#). and Norður Þ ingeyjarsýsla (Figure #$.+). The area can conveniently be divided into four main geographical While there has been an increase in the use of aerial units. Two of them (Kelduhver! and Mývatnssveit) are photographs and other remote sensing in recent within the so-called neo-volcanic zone characterised years, these techniques are still not routinely used in by extensive postglacial lava !elds and .at landscape archaeology in Iceland. Aerial sources have been used interrupted by Pleistocene ridges of hyaloclastite, ad hoc for both research and heritage management, indicative of subglacial eruptions. The Mývatn area and there are several projects conducted in the last (Mývatnssveit) is dominated by Lake Mývatn, a +)km# few years that have been laying the foundation for shallow eutrophic lake about *$km from the coast, more routine use. These have developed best practices surrounded by dry lava !elds on one side and extensive and demonstrate the bene!ts for archaeology and wetlands on the other. Modern day farms are situated landscape studies. A few of these projects are discussed around the lakeshore and around fertile wetlands on below. the River Kráká delta to the south. Kelduhver!, down by the coast, is dominated by .at lava !elds originating A project on the history of the human habitation of in the shield volcano of Þeistareykjabunga to the south the Skaga0örður region in northern Iceland used and on the north side by sandur plains deposited by low altitude oblique aerial photography to register the glacial River Jökulsá á Fjöllum. The modern day the deserted farmsites in the most remote valleys farms lie on the border between the lava and the (Pálsson '"""–#$'$). Another project on the early sandur plain, but a large number of ancient deserted arable cultivation in Iceland made extensive use of farms lie on the lava !eld to the south. conventional and infra-red aerial photographs, as well as other sources, to demonstrate the presence or absence A third landscape area is the valley and moorland of ancient sites with cultivation remains, and to study between Lake Mývatn and the coast. This is outside the discontinuity of arable cultivation in relation to the the zone of present day volcanic activity although deteriorating climate from the '%th century onwards big prehistoric lava streams did .ow down two of the (Guðmundsson et al. #$$%). In addition, another project, valleys. The bedrock is moraine-covered interglacial conducted in #$$) by an undergraduate student at the basalt carved by ice and with a rather thin layer of University of Iceland, assessed the utility of high altitude organic soil. The shallow valleys are orientated north- vertical aerial photographs for archaeological survey. south and provide the focus for most of the present The photographs were viewed at a magni!cation of day farms. They are separated by ridges of moorland '$x and while new features were mapped, the results that extend like !ngers from the highland plateau to showed some limitations. For example, mainly linear the south. The fourth landscape area is the peninsula and larger features such as areas of peat cutting and of Tjörnes that lies between Kelduhver! and Húsavík. boundaries were added to the records, but the ability to This area has a low rocky coastline with small streams at recognise smaller and more discrete sites was limited by regular intervals, and at present only the western half is the photograph scale and resolution (Sveinbjarnarson inhabited. The bedrock is the same as described before #$$)). In another project, four landscape areas covering but with thick banks of raised marine sediments on the about #$$km# were mapped using DigitalGlobe, pan- coast. sharpened Natural colour and DRA contrast enhanced $.-m pixel resolution satellite imagery, captured The whole habitable area of northeast Iceland was between #$$# and #$$- (Lárusdóttir & Aldred #$$,). apparently covered with birch (Betula pubescens) These projects demonstrated the relative ease of both scrubland at the time of !rst settlement. Today it creating and using aerial and satellite sources to expand is almost devoid of woodland. The vegetation on and enhance our existing knowledge of archaeology, the moorland and lava areas, and in some of the which can be further illustrated in the main case study uninhabited parts of the valleys, is mostly heath-like of this paper – the ongoing work in northeast Iceland. (Nielsen '""*), dominated by dwarf birch (B. nana) and #%- EAC OCCASIONAL PAPER NO. *

Figure #$.+: The study area in northeast Iceland.

crowberry (Empetrum nigrum). The inhabited parts accommodating tenant farms within their territory. of the study area have much more grassland. Hay for Wildlife resources include !sh such as salmon, trout winter fodder is the only substantial crop produced and Arctic charr in the lakes and rivers and abundant on the hay!elds around the farms. Bogs and other sea !sh, mostly cod. Rock ptarmigan, Arctic fox and wetlands are scattered throughout the area except in gyrfalcon occur in the dry upland and coastal areas, and the largest lava !elds. Extensive blanket and string bogs seabirds (and their eggs) on the coastal islands. Duck are a characteristic of the southernmost moorlands. eggs are utilised extensively in Lake Mývatn and there The soil is minerogenic and rather thin (commonly is a big eider colony on the coast. All these resources about 'm). The soil cover is mostly continuous but is were utilised in the Viking period, and the coastal locally eroded in many exposed and steep places. A harvest was brought inland (McGovern et al. #$$-). massive erosion front migrating from the highlands reached the area south and east of Lake Mývatn in the Categories of archaeological features ')th – ',th century and another erosion front is active in Although the main task of the project was to map the the area northwest of Lake Mývatn (Hólasandur). extensive wall system, the aerial observations have also registered a large number of other archaeological The study area has been inhabited from the very features. Below are some of the main characteristics of beginning of settlement and like elsewhere in Iceland the archaeology derived from the aerial surveys. Some was based on dispersed single-household farms, often of the listed features below date from the Viking period. #$ The archaeological landscape of northeast Iceland: a ghost of a Viking Age society #%)

Walls The ancient turf walls are the most prominent feature collapsed long before a characteristic tephra layer from of the archaeological landscape (Figure #$.%). They AD '%)) was deposited. The focus of the study has been run long distances, criss-crossing the moorlands on the out!eld walls, less on in!eld ones encircling the and heaths (a total of about %$$km at the last count; hay!elds which may have a more complex history due Einarsson et al. #$$#; Aldred #$$,). They seem roughly to wall building activities associated with the ',th – '"th contemporaneous, though there are indications of century agricultural reformation. multiple phases of construction in some areas (e.g. Figure #$.*). For the most part, a basic pattern can be The walls form a basic pattern of home range enclosures discerned (Figure #$.-), modi!ed by only minor repairs, (Figure #$.%), each one corresponding to an individual rebuilds and additions, and it appears that the walls farm but subdivided for more local management of went out of use a few generations after they were built. grazing, stock manipulation and protection of growing Dating of some, using tephra (volcanic ash) from #) crops. In the valley landscape the walls run uphill on the trenches to determine the latest and earliest times a probable boundaries between neighbouring farms. A wall may have been constructed or fallen out of repair, horizontal wall on the hillside divides the rangeland places the majority in the '$th to ''th centuries. Their above the farm into a near and far section. It is uncertain maintenance was discontinued sometime before the if the horizontal wall marks a limit of ownership or is just '+th century and all the dated walls, except one, had conveniently placed for stock management. The walls,

Figure #$.%: A map of the study area in northeast Iceland showing the extent of the medieval wall system. #%, EAC OCCASIONAL PAPER NO. *

Figure #$.*: An ancient farm site with walls at Höskuldsstaðir at Fjótsheiði.

enclosing the farm on three sides (the river typically challenge to get it mapped was a major impetus for closes the boundary on the downhill side), look like the aerial surveys presented in this paper. After trying they have been built as a single entity. The horizontal several methods for e(ective mapping we decided that wall usually joins similar ones at the neighbouring none of them was working successfully on its own. We farms, creating a continuous structure that fences ended up with a combination of methods involving entire valleys from the surrounding hills. both oblique and vertical photographs (the latter being the standard source for geodetic purposes) Most of the walls have collapsed and can only been and !eld walking. The oblique photos turned out to seen as low earthworks in the landscape. What makes be essential for the detailed interpretation of the them prominent, especially from the air, is that the walls in the landscape and often also to ascertain their collapsed wall is very broad (commonly %–-m across) absence. and that the ditches on each side, from where the turf had been dug, are still quite deep (often expanded by Small enclosures erosion) and their vegetation di(ers from that on top of A variety of small enclosures can be seen from the air, the collapsed wall. The result is that from the air many but two particular types deserve attention because of the walls look like huge wheel tracks (Figure #$.)). of their well de!ned geometry. One type is square on plan, measuring about #$m across and usually located The wall system is very extensive, and clearly the most on the in!eld side of the wall system, sometimes extensive archaeological phenomenon in Iceland. The isolated, sometimes combined with the walls and

Figure #$.-: A close-up of the medieval turf wall system at Fljótsheiði. #$ The archaeological landscape of northeast Iceland: a ghost of a Viking Age society #%"

Figure #$.): Walls in Laxárdalsheiði by the river Laxá.

can then be assumed to be contemporaneous. These of charcoal pits scattered between Lake Mývatn and square enclosures can for example be seen at the farm the coast. They tend to occur in clusters on low ridges sites of Narfastaðir, Ingiríðarstaðir, Einarsstaðir and or .at slopes not far from major horse tracks. Individual Brettingsstaðir (see Eldjárn ('",') for similar structures pits within a cluster are spaced some *$–'$$m apart. in Skíðadalur, North Iceland). An archaeological Sometimes, the pits occur in pairs. Most are about #m excavation underway at Ingiríðarstaðir has discovered across on the surface (max. %m) and about *$cm deep ard-marks inside one of the enclosures (Howell Roberts and look square-shaped from the air. pers. com.). The other type is represented by perfectly circular enclosures, -–'#m in diameter. Further Excavations reveal an original circular outline and the studies are needed to determine if these two types of squarish shape must be a secondary feature. On the enclosures represent functional groups or not. ground a raised rim of soil upcast is clearly visible, but this is less apparent from the air (Figure #$.,). Nine Charcoal pits pits in one cluster have been dated, and all fall within The natural climax vegetation in most of the study the period AD '$$$–'#$$, except one that is slightly area is birch woodland. Most of this disappeared rather younger, but still dates before '+$$ (Church et al. quickly after settlement (landnám), or in less than a #$$)). Charcoal was produced throughout the history century in the south of Iceland but more gradually in of Iceland, up to the middle of the #$th century. The the west and north, including our study area (Hallsdóttir product was used locally for smithy work (including '",); Lawson et al. #$$); Lawson #$'$; Gathorne-Hardy scythe sharpening), and in medieval times also for the et al. #$$"). Our aerial surveys have disclosed hundreds smelting of bog iron.

Figure #$.,: Charcoal pits on Laxárdalsheiði. #*$ EAC OCCASIONAL PAPER NO. *

Horse tracks Hay stores Overland in Iceland has taken place either on Early #$th century cultivated hay!elds only provided foot or on horses, and wheeled transport did not exist about half of the total hay production in the study until the late '"th century when the !rst roads were area (Hólmgeirsson '"),). Most of the other hay came made for wheeled carriages. Most road construction from wetlands. According to '"th and #$th century was in the form of ‘bridges’, i.e. short dykes of sod ethnographic sources, the hay was stacked on raised or stone across rock clefts or marshes. Travellers on ground on the edge of the marsh or, if the marsh horseback kept to traditional routes that tended to was large, on specially built platforms (in Icelandic = follow the contours of the landscape. Aerial surveys heystæði), until it could be transported by sledge in easily identify horse tracks. The hooves of the horse cut winter. These hay platforms, square or oval in shape, quickly through the topsoil and soon a deep track was were used extensively until the mid #$th century and formed and then parallel ones when the earlier tracks are clearly visible from the air (for example, south of became too deep to use. The most travelled routes Lake Mývatn, or on the islands within the Lake Mývatn). thus developed multiple tracks, with tens or even a None of them has been dated but it is not unlikely that hundred furrows in a braided spread across a wide area their use goes back to medieval times. The hay was (Figure #$."). There are two types of tracks inscribed covered by turf strips for protection. The remains of the into the landscape. Inter-region tracks allowed for the turf tended to accumulate at the edge of the platform, movement of goods and people between a resource forming a rim that can easily be mistaken for a house area and a distribution centre, while intra-region tracks ruin. connect individual settlements, as well as activity areas within farmland. The intensity, as well as duration of The medieval law books frequently mention enclosures use, is represented by how deeply incised the track is, for storage of hay (Icelandic = heygarður). No structures as well as its breadth. Fainter traces of tracks, perhaps have been identi!ed from the air that !t their less used rather than any later in date, are evident in description, but it seems likely that such hay enclosures many areas. Occasionally tracks run along the line of were small in!eld features and attached to other wall walls or on top of them. structures.

The horse tracks are obviously an accumulation of over Herding structures ',$$$ years of transport history and some may retain Some of the walls incorporate variously shaped their medieval period locations and some may not. enclosures that may be related to stock management. Many of the tracks seem to postdate the walls as they Ancient documents, including Grágás, the medieval law appear to cut across them. Interestingly, some tracks code of Iceland, refer to pens for herding (in Icelandic can still be seen leading to farms that were deserted = rétt) but their design is not described and they most centuries ago. The most impressive horse tracks in the likely varied with both purpose and landscape. None study area are south of the trading harbour of Húsavík of the observed structures resembles the modern and branches of this massive track can be traced into sheep folds used for managing sheep driven from all the main valleys and all the way to the Lake Mývatn the upland commons in the autumn. The date of the region. Some of the tracks may relate to transport of early medieval phase of the herding structures has sulphur from the areas east and southeast of Lake yet to be demonstrated archaeologically, but it can Mývatn to Húsavík and perhaps Gásir by Eyja0örður for be intimated that they were similar to those that have export to continental Europe. Sulphur was mined in this been excavated; dating to the early-')th century (Aldred region over seven to eight centuries. In one large area, #$'$). Hólasandur between Mývatn and Húsavík, the topsoil has been blown away and the underlying sandy subsoil In the Mývatn area, the earliest sheep folds seem to be has been exposed. Here the main track expands into located away from the settlements in the grazing areas, a -–"m broad road where rocks have been cleared to both to the south in Suðurafrétt, and in Norður0öll the side to create a soft substrate for riding. The age of (Gæsadalur and east of Hágöng), but are also evident this unique construction is unknown but it is probably in other places across Iceland (Aldred & Madson #$$"; post-medieval, judging from preliminary dating of the Aldred #$'$). The earliest structures are simple in their erosion (unpublished data). form (c. *$$m#), or utilised topography, rather than the multi-compartment folds of '"th century date and later.

Churches or chapels In the beginning of Christianity in Iceland churches were private property and most were built close to farmhouses. The church or chapel was oriented east- west and enclosed within a circular wall that was both symbolic and practical (to keep animals away). Most churches were small and the diameter of the circular wall was only #$–+$m. Square churchyards became the norm in the '"th century (Jónasson '"-', +%)). Only four ancient (but undated) churches/chapels have been

Figure #$.": Horse tracks by Lake Mývatn. #$ The archaeological landscape of northeast Iceland: a ghost of a Viking Age society #*'

Figure #$.'$: Brettingsstaðir, an ancient farm site with a church or chapel (inside the larger circle) and a small circle with unknown function. Remains of a #$th century farmhouse above and an apparently Viking Age longhouse below the circular walls.

seen from the air in our study area: at Ingiríðarstaðir and Although the assemblies were the stage for many epic Einarsstaðir in Þegjandadalur valley, Brettingsstaðir in events in the Saga literature, surprisingly little is known Laxárdalur valley (Figure #$.'$) and Saltvík by Húsavík. about their spatial organisation, and there is no clear A church at Hofstaðir was located through a ground typology to support aerial observations (cf. Vésteinsson resistance and magnetometer survey (Horsley & et al. #$$%; see also Friðriksson '""%). A cluster of small Dockrill #$$#) and a few other medieval churches are and evenly spaced house ruins in protected locations known from historical sources. close to water may qualify as possible assembly sites. Some sites with these characteristics correlate Assembly sites with locations given in the ancient literature and are Assemblies (Icelandic = þing) were an important part of sometimes supported by place name evidence. The the social organisation from the very early settlement visible layout of the only documented þing-site in our in the late "th century and were formalised on a study area, Þingey, di(ers in having a series of house nationwide scale in AD "+$ when the general assembly ruins that seem connected gable to gable to form at Þingvellir was established. It is normally assumed two parallel rows. Extra complexity is added to the that the whole assembly structure collapsed along with archaeology of the Þingey site because of '"th century the judicial system in the '+th century, to be replaced by farming activities. a more centralised judicial system and royal executive power in the wake of Iceland’s union with Norway in The nearby site of Skuldaþingsey, however, has a '#-#. dispersed cluster of +$ ruins (Figure #$.'') that can only

Figure #$.'': An assembly site at Skuldaþingsey where #- of the +$ booths can be seen. #*# EAC OCCASIONAL PAPER NO. *

Figure #$.'#: Fremri Fjöll, a major farm site in Kelduhver!. Top: The large house from the air. Bottom: A line drawing based on detailed measurements of the same house and another longhouse close by. be interpreted as booths for some kind of assembly The lack of clear typology exempli!ed by the striking (an interpretation also assisted by the þing element in dissimilarity of the two major and adjacent sites Þingey the name). The location of this large assembly site so and Skuldaþingsey makes it di/cult to assign a function close to the traditional major site in Þingey is somewhat to two other suspected assembly sites, Leiðarhóll by puzzling. Both assembly sites date to the medieval Helgastaðir and Leiðarnes by the River Fnjóská. The period and an excavation in Skuldaþingsey shows names suggest minor assemblies (Icelandic = leið refers evidence of prolonged but seasonal use (Vésteinsson to a local assembly for the announcements of decisions et al. #$$%). made at regional assemblies). The Leiðarhóll site is also #$ The archaeological landscape of northeast Iceland: a ghost of a Viking Age society #*+ complicated by later farming activities – in fact there walls, a couple of churches and a Viking period grave is nothing assembly-like about it other than the name. !eld. The other site, Brettingsstaðir, is a well-preserved The Leiðarnes site, however, certainly quali!es as an farm site in the valley of Laxárdalur and has what seems assembly site, judging by the name and by the typology to be a double Viking period hall (Figure #$.'$), a church discussed above. It has a group of evenly spaced ruins or chapel ruin, some outhouses and a system of walls. of similar size, located on a peninsula created by a The ridge above Brettingsstaðir has a large cluster of meander in the river. Both sites are awaiting aerial charcoal pits. recording. The farm sites in the marginal and usually better Farm sites preserved settlements are of various types but most The set of ancient farmsteads observed from the air is share the feature of one or more quasi-circular or sub- certainly a biased sample, as the best preserved sites rectangular walled enclosures (Figure #$.'+). If there are are the marginal ones that were abandoned early. multiple rings of enclosure they are often concentric, The '#–'+th century saw large-scale abandonment of and frequently connected by transverse walls. The the marginal, mostly upland and interior settlements house complexes are usually in the innermost circle. The (Thorarinsson '")-; Rafnsson '""$; Sveinbjarnardóttir district of Kelduhver! has an impressive concentration '""#). Although farm abandonment may have taken of this type of settlement and some of those are truly place in other periods, most of the farms that survived marginal in every sense, like those at Bláskógar with the medieval period were still in use in the early #$th three large-diameter quasi-circular fences adjacent to century and the ruins of the original settlement have each other in the mid-slope lava !elds of the shield been superseded by modern development. Also, the volcano of Þeistareykjabunga. marginal upland farm sites will tend to have lower soil accumulation rates, and therefore be more visible, than The origin of multiple concentric quasi-circular those of the richer lowland. fenced enclosures is not understood, but it has been suggested that they represent stages in an expanding There are two notable exceptions where major farm sites hay!eld from the time of !rst settlement (Líndal '"*'). have survived. One is the well known archaeological Alternatively they may re.ect a contracting hay!eld in site of Hofstaðir by Mývatn, a large '$th century farm response to deteriorating environmental conditions, with an oversized Viking type longhouse, showing multiple periods of settlement and abandonment, and evidence of ritual feasting (Lucas #$'$). The wall system !nally, and perhaps most commonly, they may be a associated with the farm is also well preserved. The way of subdividing the home range for management other is Fremri Fjöll in the Kelduhver! district. It has not purposes. been dated but the ground plans of the houses suggest "th–''th century (Christian Keller pers. com.). The site is Many of the well preserved abandoned farm sites beautifully preserved and the constructions are still show signs, sometimes backed up by historical prominent above ground with two exceptionally large documentation, of much later use as grazing stations. halls (Figure #$.'#), a set of at least seven smaller houses If this later activity is within the last one hundred and a complex system of walls. years or so the fertilising e(ect of the activity is easily recognizable from the air by a rich growth of grass. Two areas deserve special mention because of good The vegetation on older ruins conforms with the preservation. One is the valley of Þegjandadalur that natural vegetation, usually dwarf shrubs. It is possible was deserted early (probably long before the '%)) that a site has oscillated between permanent and tephra) and has well-preserved farmhouse complexes, seasonal use in response to changing population size

Figure #$.'+: A probably medieval deserted farm site with a triple enclosure by the River Reykjakvísl. There is a riding path to the right. The yellow colour inside the innermost enclosure is indicative of farming activity in modern times. #*% EAC OCCASIONAL PAPER NO. *

Figure #$.'%: A peat pit in Fljótsheiði. or environmental conditions. Much work remains to L-shaped internal pro!le (Figure #$.'%). If the wetland is correlate the surface appearance of the various houses situated on a slope there is usually a narrow cut out of to their original structure and function. the pit for draining. None of the pits has been dated, but some were used till the mid-#$th century. Single houses Quite a few ruins of single houses have been located from the air. Some are very small (% by %m) and occur Discussion along the main riding paths. These may be shelters for travellers (Icelandic = sæluhús) or shepherds (Icelandic Cultural value of the monuments and their protection = smalako#), shielings (summer grazing stations) or Clearly, the spatial extent of the northeast landscape, temporary stores of bog iron or peat, just to mention a its good preservation, as well as the antiquity and few possibilities. importance of its archaeology for Iceland, if not Europe, calls for a policy of protection and other management. Pre-Christian graves There are several challenges that need to be met in Christianity was made the o/cial religion in AD '$$$ order for the archaeological landscape to be protected and in the approximately '+$ years before that non- or otherwise managed to ensure its cultural value Christian Viking Age burial customs were prevalent. We for future generations. The northeast landscape, like have been able to see a number of looted pre-Christian many areas where archaeology is prevalent, is both a graves from the air. Looted graves have a telltale scar relict and a living landscape. The !rst challenge is to and signs of upcast soil, even if the looting dates to the !nd ways in which the demands of the relict and the Middle Ages (Vésteinsson #$$%b). Untouched graves living can be managed to complement each other. have no known external features that allow their The extensive character of the archaeology, especially location either at ground level or from the air. the wall systems, also presents another challenge of how best to protect or manage a resource that is so Peat pits widely distributed. Clearly, there is a need to be well Peat was used as fuel along with wood from an early informed, and to continue to monitor the resource for date (Simpson et al. #$$+). The peat was cut in certain any potential or active threats, such as by development peat pits (locally known as ‘svarðargra#r’) in the associated with infrastructure projects, agriculture, wetlands and dried on nearby hillocks before transport. a(orestation, as well as natural threats such as volcanic The pits are easily recognisable from the air by their activity or erosion. Providing the opportunity for squareness, but they vary in size and shape. Clusters protection and management that is connected with of small (less than *m in diameter) pits are frequent. both natural and cultural factors, and archaeology’s Larger pits (over '$m across) have a characteristic relationship to both, is an important challenge to meet. #$ The archaeological landscape of northeast Iceland: a ghost of a Viking Age society #**

A clear road-map for protection and management lies Mývatn with three well preserved medieval farm in two complementary bodies of legislation – !rstly sites (Brenna, Selás and Vindbelgur), shielings Icelandic heritage law (e.g. Law number '$)/#$$' (summer grazing stations) and a pagan grave !eld; Þjóðminjalög) and second, European frameworks, +. Fljótsheiði with an extremely long wall and such as the European Landscape Convention (ELC, side walls and associated remains of medieval Council of Europe #$$$; ESF Science Policy Brie!ng %' buildings; #$'$). Both provide a strategic policy to examine how a %. Hvammsheiði moorland with the farm sites of landscape, such as the one described in this paper, can Litlu Núpar and Íragerði and a few massive walls be protected and managed in a way that accounts for crossing the moorland, also a boat burial; both the historical landscape and the dynamic living *. The concentration of deserted encircled farms one occupied by people depending on it for their above the main row of present day farms in livelihood. Kelduhver!, including the extraordinarily well preserved site of Fremri Fjöll; Knowledge is a prerequisite for e/cient management -. The well preserved system of walls immediately of the historic landscape. Gaps in our knowledge need south of Húsavík, including a church, many house to be identi!ed and !lled, and a monitoring programme ruins and a pagan grave !eld (cf. Lárusdóttir #$$)); of the status of the landscape should be established, ). The archaeological hotspot of Bakki just north of such as Historic Landscape Characterisation (Aldred & Húsavík with a dense complex of walls, irrigation Fairclough #$$+; Fairclough & Macinnes #$$+; Fairclough structures and peat pits; #$$)). These monitoring programmes should be ,. Seljadalur, a deserted valley with a row of well- conducted with the aid of di(erent stakeholders, preserved farm ruins with an impressive system of including landowners. walls; ". Þingey and Skuldaþingsey, the well preserved dual While direct protection is perhaps suited to exceptional assembly site; landscapes, such as the ones listed below, the majority '$. The sulphur mining area and farm site of of the landscape has no such protection and needs to Þeistareykir. be e(ectively managed. Protection alone, therefore, leaves some ambiguity over areas not a(orded special The list above includes areas where pre-modern protection, and one could view this as a rather passive landscapes remain more intact than elsewhere and form of heritage management that is subject to covers some of the hotspots of the Viking period contemporary economic and socio-political conditions society. Of course there is a multitude of more recent (Aldred & Friðriksson #$$,). A management approach, archaeological monuments that deserve attention however, recognises the value of all landscape, as and should in.uence any plans for regional scale indicated through the ELC framework, and a (pro)active management. These include the extensive (undated) approach to land management. Direct protection walls built of lava rock by Lake Mývatn and along the should then be seen as an extreme and localised case River Laxá, and the road on Hólasandur. The various of more wide-ranging management. structures associated with sheep farming, folds and pens for milking sheep and feeding lambs many The mapping of archaeology from aerial sources, of which are integrated in the lava landscape in an combined with !eld based archaeological survey, interesting way, should also be included. Unique to the and small-scale excavation designed to date and Mývatn area is the extensive use of lava caves as retreats characterise speci!c parts of the resource, give for sheep from the bloodsucking black.y (Simulium). the north-eastern landscape a good basis for the development of an exemplar for best practice of General remarks landscape management and protection in Iceland. It is rare to have a large scale, one thousand year old archaeological landscape that is so well preserved, Much work is needed to classify the archaeological as well as one that was moulded within such a short landscape in terms of preservation or management period and still shows the pattern of landholdings, value. Nevertheless, our aerial surveys already allow resources and !ne scale management. The potential us to pinpoint ten outstanding areas in this respect. for using the wall system to analyse the settlement They have a high density of well-preserved and clearly pattern in relation to landscape characteristics is visible remains, often interconnected by a system of enormous. We not only see the pattern of fences but walls and horse tracks, in an attractive landscape that is also the communication routes and the distribution of not interrupted by large-scale modern activities. some resources like peat and woodland. Also, religious centres and other assembly sites can be pinpointed. This Those ten areas are: information is augmented by detailed archaeological and palaeoecological studies of individual sites yielding '. Þegjandadalur, a deserted valley with a multitude a good record of diet, fuel and trade (Lucas #$'$). of medieval farms with a complex wall system, two churches and a large pagan grave !eld The geography of any human society is the result of (Hreiðarsdóttir & Roberts #$$"); dynamic processes that depend on the productivity, #. The upper part of Laxárdalur, including the area dispersion and seasonality of resources and also between Lake Mývatn and Laxárdalur, with on population pressures and a range of cultural highlights including Hofstaðir, Brettingsstaðir responses. The Icelandic farming society has always and an unnamed farm #km north thereof, all with been a dispersed territorial system based largely on well preserved wall systems, the west side of Lake an exclusive home range for summer grazing of sheep #*- EAC OCCASIONAL PAPER NO. * and cattle and haymaking for winter fodder (see e.g. References Vésteinsson '"",, #$$$, #$$*). Most farms also had Aldred, O. #$$,: Unfamiliar landscapes: in!elds, outstations for summer or winter grazing and access to out!elds, boundaries and landscapes in Iceland. upland commons. Territoriality and animal husbandry In Chadwick, A.M. (ed.): Recent Approaches to the stimulates the erection of fences, given the capacity Archaeology of Land Allotment. BAR International to invest in such structures. Fences are needed to Series ',)*, Archaeopress, Oxford. #""–+#'. keep your animals within reach and those of your neighbours at bay. The home range may be subdivided Aldred, O. #$'$: Réttir in the landscape. Archaeological for more local management of species, sexes and age investigations of sheep-folds in Skútustaðahreppur groups and also to protect crops or to manure the and other neighbouring districts. Unpublished ground. The con!guration of the walls and the large report, International Polar Year. investment involved must therefore tell us a great Aldred, O. & Fairclough, G. #$$+: Historic landscape deal about the geography, structure and resources characterisation: Taking stock of the method – The of the Icelandic society of the Viking period (Aldred National HLC Method Review $%%$. English Heritage, #$$,). The wall con!guration shows evidence of being London. in.uenced mainly by landscape characteristics and population density. Tightly packed home ranges tend Aldred, O. & Friðriksson, A. #$$,: Iceland. In Fairclough, to have straight boundaries and polygonal shapes. G. & Møller, P.G. (eds): Landscape as Heritage. The Linear landscapes, like narrow valleys or coastlines, management and protection of landscape in Europe, tend to have a single row of square enclosures, while a summary by the COST A$& project ‘LANDMARKS’. home ranges on a .at terrain tend towards circular or COST, University of Berne, Institute of Geography polygonal shapes. (G)"), Bern. '%*–**. Aldred, O., Hreiðarsdóttir, E.Ó. & Sveinbjarnarsson, The archaeological remains in the study area may hold Ó.G. #$'$: On the precipice: aerial archaeology in the key to understanding the dramatic changes that Iceland. Archaeologia Islandica ,, '''–#'. clearly occurred all over Iceland in the '#th–'+th century and involved large scale contraction of the settlements, Aldred, O. & Madson, C. #$$": Réttir in the landscape. perhaps preceded by the disuse of the wall system. A study on the interactions between humans Hypotheses as to the causal factors include landscape and animals through sheep-fold monuments. degradation, climatic cooling and an economic shift International Polar Year unpublished report. (see e.g. Dugmore et al. #$$$, #$$*). None of these Bragason, Þ. & Guðmundsson, M. '",,: Fifty Years of hypotheses are mutually exclusive, but they involve Change and Development: Aerial Photographs from either migration of people due to shifting resources Iceland. Hörpuútgáfan, Landmælingar Íslands. or the decimation of the population by high mortality rates. Either way, they concern the interaction between Church, M.J., Dugmore, A.J., Mairs, K.A., Millard, people and environment, and the dynamics of the A.R., Cook, G.T., Sveinbjarnardóttir, G., Ascough, change may be important not only for our perception of P.A. & Roucoux, K.H. #$$): Charcoal production the but also that of other settlements during the Norse and early medieval periods in in the North Atlantic. Eyja0allahreppur, southern Iceland. Radiocarbon %", -*"–)#. Council of Europe, #$$$: European Landscape Acknowledgements Convention. Council of Europe (Treaty series No. ')-), Florence. The Medieval Wall Project, of which the aerial survey was part, is a joint undertaking of a group of people Dugmore, A.J., Newton, A.J., Larsen, G. & Cook, from the Institute of Archaeology, Iceland, the Myvatn G.T. #$$$: Tephrochronology, environmental Research Station and the University of Oslo. Apart change and the Norse colonization of Iceland. from the present authors they are: Birna Lárusdóttir, Environmental Archaeology *, #'–+%. Christian Keller, Elín Hreiðarsdóttir, Oddgeir Hansson, Dugmore, A.J., Church, M.J., Buckland, P.C., Edwards, Orri Vésteinsson and Stefán Ólafsson. We would like K.J., Lawson, I.T., McGovern, T.H., Panagiotakopulu, to thank our pilots Leifur Hallgrímsson and Birgir E., Simpson, I.A., Skidmore, P. & Sveinbjarnardóttir, Steingrímsson, and also our many !eld assistants for G. #$$*: The Norse landnám on the North Atlantic hard work during ground surveys and excavations. islands: an environmental impact assessment. Polar The project was supported by the Icelandic Centre for Record %, #'–+). Research (RANNÍS), Vegagerðin (the Icelandic Road Administration) and Þjóðhátíðarsjóður (a fund that Einarsson, Á., Hansson, O. & Vésteinsson, O. #$$#: commemorates the "th century settlement in Iceland). An extensive system of medieval earthworks in Finally we thank Orri Vésteinsson for critical reading of NE-Iceland. Archaeologia Islandica #, -'–)+. the manuscript. All photos are by the !rst author. Eldjárn, K. '",': Uslaréttir. Árbók Hins íslenska fornleifafélags '()%. Reykjavík. '$'–'$. Fairclough, G. #$$): The contemporary and future landscape: change & creation in the later #$th century. In McAtackney, L., Palus, M. & Piccini, A. (eds): Contemporary and Historical Archaeology in Theory, Paper for the $%%* and $%%+ CHAT conferences #$ The archaeological landscape of northeast Iceland: a ghost of a Viking Age society #*)

(Studies in Contemporary and Historical east Iceland. Institute of Archaeology Monograph Archaeology %, BAR International Series '-))). Series ('), Reykjavík, #-–*%. Archaeopress, Oxford. ,+–,. Lawson, I.T., Gathorne-Hardy, F.J., Church, M.J., Fairclough, G. & Macinnes, L. #$$+: Understanding Newton, A.J., Edwards, K.J., Dugmore, A.J. & Historic Landscape Character, Topic paper ,, Einarsson, Á. #$$): Environmental impacts of the Landscape Character Assessment Guidance for Norse settlement: palaeoenvironmental data from England and Scotland. Countryside Agency, Scottish Mývatnssveit, northern Iceland. Boreas +-, '–'". Natural Heritage, Historic Scotland and English Líndal, J.H. '"*': Um forn mannvirki og örnefni Heritage. á Lækjamóti í Víðidal. Árbók hins íslenzka Friðriksson, A. '""%: Sagas and Popular Antiquarianism fornleifafélags '"%"–*$, ),–'$'. in Icelandic Archaeology. Aldershot, Avebury. Lucas, G. (ed.) #$'$: Hofstaðir. Excavations of a Viking Friðriksson, A. & Vésteinsson, O. '"",: Fornleifaskráning Age feasting hall in North-east Iceland. Institute of – Brot úr íslenskri vísindasögu. Archaeologia Archaeology Monograph Series ('), Reykjavík. Islandica ', '%–%%. McGovern, T.H., Perdikaris, S., Einarsson, Á. & Sidell, Gathorne-Hardy, F.J., Erlendsson, E., Langdon, P.G. & J. #$$-: Coastal connections, local !shing, and Edwards, K.J. #$$": Lake sediment evidence for late sustainable egg harvesting: patterns of Viking Holocene climate change and landscape erosion Age inland wild resource use in Mývatn district, in western Iceland. Journal of Paleolimnology %#, Northern Iceland. Environmental Archaeology '', %'+–#-. ',)–#$*. Guðmundsson, G., Snæsdóttir, M., Simpson, I., McGovern T.H., Vésteinsson, O., Fridriksson, A., Hallsdóttir, M., Sigurgeirsson, M.A. & Árnason, Church, M., Lawson, I., Simpson, I.A., Einarsson, Á., K. #$$%: Fornir akrar á Íslandi. Árbók hins íslenzka Dugmore, A., Cook, G., Perdikaris, S., Edwards, K.J., fornleifafélags (#$$#-+), )"–'$-. Thomson, A.M., Adderley, W.P., Newton, A., Lucas, G., Edvardsson, R., Aldred, O. & Dunbar, E. #$$): Hallsdóttir, M. '",): Pollen Analytical Studies of Human Landscapes of settlement in northern Iceland: In-uence on Vegetation in Relation to the Landnám Historical ecology of human impact and climate Tephra Layer in Southwest Iceland. Lund University, .uctuation on the millennial scale. American Department of Quaternary Geology (Lundqua Anthropologist '$", #)–*'. Thesis ',), Lund. Nielsen, Ó.K. '""*: Karrar og gróðurfar. (English Hólmgeirsson, G. '"),: Fóðurbirgðaskoðun í summary: Rock ptarmigan density and heathland Reykdælahreppi ,.–##. mars '"'". Árbók Þingeyinga vegetation in Suður-Þingeyjarsýsla, NE-Iceland). ('"),), '*"–--. Náttúrufræðingurinn -*, ,'–'$#. Horsley, T.J. & Dockrill, S.J. #$$#: A preliminary Ólafsson, G. '""': Fornleifaskráning og fornleifavernd. assessment of the use of routine geophysical Sveitarstjórnarmál II, ,$–#. techniques for the location, characterisation and interpretation of buried archaeology in Iceland. Pálsson, H. '"""–#$'$: Byggðasaga Skaga/arðar. Vols. Archaeologia Islandica #, '$–++. '–*. Sögufélag Skaga0arðar, Sauðárkrókur. Hreiðarsdóttir, E.Ó. & Roberts, H. #$$": Þögnin ro!n. Rafnsson, S. '""$: Eyðibyggð í Hrafnkelsdal og á Fyrstu niðurstöður fornleifarannsókna á eyðibyggð Brúardölum: Brot úr byggðarsögu Íslands. Rit hins á Þegjandadal. Árbók Þingeyinga #$$,, *–#%. íslenzka fornleifafélags, Reykjavík. Ísaksson S.P. & Helgason, Þ.S. '""*: Vetrarmyndir frá Science Policy Brie!ng %' #$'$ Landscape in a changing Nesi við Seltjörn og Laugarnesi. Árbók hins íslenzka world. European Science Foundation & ESF fornleifafélags ('""%). COST (Accessed '$/'#/#$'$: http://www.esf.org/ publications/ science-policy-brie!ngs.html). Jónasson, J. '"-': Íslenzkir þjóðhættir (The Ethnography of Iceland). +rd edition. Ísafoldarprentsmiðja, Simpson, I.A., Vésteinsson, O., Adderley, W.P. & Reykjavík. McGovern T.H. #$$+: Fuel resources in landscapes of settlement. Journal of Archaeological Science +$, Lárusdóttir, B. #$$): Settlement organization and '%$'–#$. farm abandonment: The curious landscape of Reykjahver!, North-East Iceland. In Davies, W., Sveinbjarnardóttir, G. '""#: Farm Abandonment Halsall, G. & Reynolds, A. (eds): People and Space in Medieval and Post-Medieval Iceland: an in the Middle Ages, *%%–'*%% (Studies in the Early Interdisciplinary Study. Oxbow (Monograph ')), Middle Ages '*). Brepols, Turnhout. %*–-+. Oxford. Lárusdóttir, B. & Aldred, O. #$$,: Kortlagning fornleifa Sveinbjarnarson, Ó.G. #$$): Aerial Survey: af gervihnattamyndum. Fornleifastofnun Íslands Archaeological survey through high altitude aerial (FS+,"–$,','), Reykjavík. photographs. Unpublished BA thesis. University of Iceland, Reykjavík. Lawson, I. #$'$: The palaeoenvironment of Mývatnssveit during the Viking age and Early Thorarinsson, S. '")-: Gjóskulög og gamlar rústir. Medieval period. In Lucas, G. (ed.) Hofstaðir. Brot úr íslenskri bygðasögu. Árbók hins íslenzka Excavations of a Viking Age feasting hall in North- fornleifafélags '(&0, *–+,. #*, EAC OCCASIONAL PAPER NO. *

Vésteinsson, O. '"",: Patterns of settlement in Iceland. Íslands (Institute of Archaeology, Iceland), A study in prehistory. Saga-Book of the Viking Reykjavík. Society XXV, '–#". Vésteinsson, O. #$$*: Archaeology of economy and Vésteinsson, O. #$$$: The archaeology of landnám. society. In McTurk, R. (ed.): A Companion to Old Early settlement in Iceland. In Fitzhugh, W.W. & Norse – and Culture. Blackwell, Ward, E. (eds): Vikings. The North Atlantic Saga. Oxford, )–#-. Washington DC, Smithsonian Inst. Press. '-%–)%. Vésteinsson, O., Einarsson, Á. & Sigurgeirsson, M.Á. Vésteinsson, O. #$$%a: Icelandic farmhouse #$$%: A new assembly site in Skuldaþingsey in excavations: !eld methods and site choices. NE-Iceland. In Guðmundsson, G. (ed.): Current Archaeologia Islandica +, )'–'$$. issues in Nordic Archaeology. Proceedings of the $'st Conference of Nordic Archaeologists, 0–( September Vésteinsson, O. (ed.) #$$%b: Archaeological studies in $%%', , Iceland. Society of Icelandic Saltvík $%%*. Unpublished report, Fornleifastofnun Archaeologists, Reykjavík, ')'–".