stable isotopes, chronology, and bayesian models for the...

Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=uica20

The Journal of Island and Coastal Archaeology

ISSN: 1556-4894 (Print) 1556-1828 (Online) Journal homepage: https://www.tandfonline.com/loi/uica20

Stable Isotopes, Chronology, and Bayesian Modelsfor the Viking Archaeology of North-East Iceland

W. Derek Hamilton & Kerry L. Sayle

To cite this article: W. Derek Hamilton & Kerry L. Sayle (2019) Stable Isotopes, Chronology,and Bayesian Models for the Viking Archaeology of North-East Iceland, The Journal of Island andCoastal Archaeology, 14:1, 71-81, DOI: 10.1080/15564894.2017.1363097

To link to this article: https://doi.org/10.1080/15564894.2017.1363097

© 2017 The Author(s). Published withlicense by Taylor & Francis© W. DerekHamilton and Kerry L. Sayle

Published online: 07 Nov 2017.

Submit your article to this journal

Article views: 404

View Crossmark data

https://www.tandfonline.com/action/journalInformation?journalCode=uica20https://www.tandfonline.com/loi/uica20https://www.tandfonline.com/action/showCitFormats?doi=10.1080/15564894.2017.1363097https://doi.org/10.1080/15564894.2017.1363097https://www.tandfonline.com/action/authorSubmission?journalCode=uica20&show=instructionshttps://www.tandfonline.com/action/authorSubmission?journalCode=uica20&show=instructionshttp://crossmark.crossref.org/dialog/?doi=10.1080/15564894.2017.1363097&domain=pdf&date_stamp=2017-11-07http://crossmark.crossref.org/dialog/?doi=10.1080/15564894.2017.1363097&domain=pdf&date_stamp=2017-11-07

DOI: 10.1080/15564894.2017.1363097

Stable Isotopes, Chronology, andBayesian Models for the VikingArchaeology of North-East Iceland

W. Derek Hamilton and Kerry L. SayleScottish Universities Environmental Research Centre, University of Glasgow,

East Kilbride, UK

ABSTRACT

This paper reviews the results of a long-term research project that usedstable isotope analyses (δ13C, δ15N, δ34S) and Bayesian mixing modelsto better model the chronology for a presumed Viking Age cemetery atHofstaðir, near Lake Mývatn in north-east Iceland. δ13C and radiocar-bon dating indicated that many of the individuals consumed a largeamount of marine protein, which results in a marine reservoir effect(MRE), making ages older than expected. In addition to the MRE,geological activity in the region has the potential to introduce massivequantities of radioactive ‘dead’ carbon into the freshwater system, re-sulting in a very large freshwater reservoir effect (FRE) that can offsetradiocarbon ages on the order of a few thousand years. The radiocar-bon dates of organisms that derive an unknown proportion of theircarbon from both marine and freshwater reservoirs are extremelydifficult to ‘correct’, or, more appropriately, model. The research notonly highlights the complexities of dealing with multiple reservoirs,but also how important it is to develop models that are temporallyand geographically relevant to the site under study. Finally, it showshow this data can be used to inform the development of chronologicalmodels for refining the dating for archaeological activity.

Keywords Iceland, palaeodiet, stable isotopes, reservoir effects, chronological modeling

Received 7 February 2017; accepted 7 July 2017.Address correspondence to W. Derek Hamilton, Scottish Universities Environmental Research Centre,University of Glasgow, Scottish Enterprise Park, Rankine Avenue, East Kilbride G75 0QF, UK. E-mail:[email protected] versions of one or more figures in this article are available online at www.tandfonline.com/uica.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 re-production in any medium, provided the original work is properly cited.

The Journal of Island & Coastal Archaeology, 14:71–81, 2019© 2017 W. Derek Hamilton and Kerry L. Sayle. Published with license by Taylor & Francis ISSN: 1556-4894 print / 1556-1828 online

71

https://doi.org/10.1080/15564894.2017.1363097https://crossmark.crossref.org/dialog/?doi=10.1080/15564894.2017.1363097&domain=pdf&date_stamp=2017-10-25mailto:[email protected]://www.tandfonline.com/uicahttp://creativecommons.org/licenses/by/4.0/

W. Derek Hamilton and Kerry L. Sayle

INTRODUCTION

Undertaking radiocarbon dating in coastaland island environments is oftentimesfraught with challenges that inland archae-ology seldom engages, most notably theneed in many cases to determine and cor-rect for a marine reservoir effect (MRE) insamples of human and animal bone usedfor radiocarbon dating. In the island na-tion of Iceland, these complexities are am-plified as the geological activity associatedwith the Mid-Atlantic Ridge makes the is-land one of the most geothermally and vol-canically active in the world. In particular,hydrothermal venting has the potential tointroduce massive quantities of radioactive‘dead’ carbon into the freshwater system,resulting in a very large freshwater reser-voir effect (FRE) that can offset radiocar-bon ages on the order of a few thousandyears.

While the internationally ratified ma-rine calibration curve (Marine13: Reimeret al. 2013) can be used with a local reser-voir correction (�R) to adjust for the effectsof eating marine-based protein, the freshwa-ter reservoir correction should be guidedby using multiple radiocarbon measure-ments on terrestrial and freshwater sam-ples from the area (and preferably time pe-riod) under consideration. However, the ra-diocarbon dates of organisms that derivean unknown proportion of their carbonfrom both marine and freshwater reser-voirs are extremely difficult to ‘correct’,or, more appropriately, model. These cir-cumstances highlight the importance of de-veloping a full understanding of the localecosystem and food web, so that individualpeople and animals can have the percent-age of their protein that is derived from thevarious reservoirs accurately and preciselycalculated.

This paper reviews the results of along-term research project that ultimatelyused light stable isotope analyses andBayesian mixing models to better modelthe chronology for a presumed Viking Agecemetery at Hofstaðir, near Lake Mývatn innorth-east Iceland (Sayle et al. 2013, 2014,2016a, 2016b). Each step of the project was

informed by the previous results and thenew questions that were often raised. Theresearch not only highlights the complexi-ties of dealing with multiple reservoirs, butalso how important it is to develop modelsthat are temporally and geographically rel-evant to the site under study. Furthermore,the findings emphasized the need to have abetter understanding of the ecology aroundLake Mývatn, as additional factors suchas sea-spray effect and midge depositionhad the ability to alter isotopic baselinesat the extreme local level, over just a fewkilometers.

GEOGRAPHICAL AND HISTORICALBACKGROUND

In 1992, the North Atlantic Biocultural Or-ganisation (NABO) was established with theintention of improving communication andcollaboration between archaeology and pa-leoecology academics interested in the col-onization of the North Atlantic region. LakeMývatn (meaning “the lake of midges” inIcelandic) lies in the north-eastern high-lands of Iceland, and, as its name suggests,is renowned for its abundant insect life.This shallow lake, located 50 km inlandand at an altitude of 277 m above sealevel (Figure 1), is a sanctuary for breed-ing waterfowl (Einarsson 2004; Gardarsson2006). As part of NABOs “Mývatn Land-scapes Project”, excavations were under-taken at the sites of Hofstaðir and Skú-tustaðir (Figure 1), and resulted in the re-gion being documented as an area of majorarchaeological importance with respect tothe settlement, or landnám, of Viking com-munities in Iceland from around AD 870 on-wards (Einarsson and Aldred 2011; Lucas2009; McGovern et al. 2007; Vésteinsson1998).

Radiocarbon dating of terrestrial ani-mal remains and tephrochronological stud-ies from various sites surrounding the lakehave shown that settlers populated the re-gion from the late ninth century (McGovernet al. 2006, 2007). The presence of theseinhabitants is thought to have had a large

72 VOLUME 14 • ISSUE 1 • 2019

Viking Archaeology of North-East Iceland

Figure 1. Map of Iceland and Lake Mývatn showing the location of the two sites discussed in thetext (Hofstaðir and Skútustaðir).

environmental impact on the area, with theintroduction of grazing livestock and rapiddeforestation (Hallsdóttir 1987) leading tosignificant soil erosion (Arnalds et al. 1997;Dugmore et al. 2005; Lawson et al. 2007;Vésteinsson et al. 2002).

GEOLOGY OF THE LAKE MÝVATN AREA

The area surrounding Lake Mývatn is vol-canic in nature, with igneous rocks of thetholeiitic series dominating the landscape.The series is split into three subsections:1) basaltic rocks, which are most abundant,comprising of picrite, olivine tholeiite andtholeiite; 2) intermediate rocks which in-clude icelandite and basaltic icelandite; and3) silicic rocks which include dacite and

rhyolite (Jakobsson et al. 2008). Porous lavafields dominate the area, leaving the surfacecharacteristically devoid of water. The lakehas two major basins, Ytriflói (north basin),which is fed by hot springs from the Ná-mafjall geothermal field, and Syðriflói (southbasin), which is fed by cold springs alongits eastern shores (Kristmannsdóttir and Ár-mannsson 2004). As groundwater springssupply most of Lake Mývatn’s water andthere is an absence of surface water in thearea for it to mix with, the chemical makeupof the water entering the lake is very stable.Before draining into the River Laxá in thewest, the geothermal waters provide LakeMývatn with plentiful supplies of silica andsulphate, whilst cooler waters deliver phos-phate to the lake (Kristmannsdóttir andÁrmannsson 2004).

73THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY


THE PROBLEMS (AND SOLUTIONS)

It has been known that radiocar-bon dating in the area around LakeMývatn (Mývatnssveit) is complicatedby the impact of marine and freshwaterreservoir effects on both human and certainfaunal remains (Ascough et al. 2007, 2010,2011, 2012), which is due to the consump-tion of both marine fish transported fromcoastal sites and local freshwater resources(e.g., freshwater fish and waterfowl) (Mc-Govern et al. 2006, 2007). The result is ahigh degree of overlap in the values for δ13Cand δ15N between animals from terrestrial,marine, and freshwater systems, making itdifficult to use just these two isotopes forunderstanding from where humans derivetheir dietary protein.

In 2011, the collagen from a wide rangeof animal bones (cow, caprine, horse, trout,charr, haddock, cod, arctic fox, pig, and var-ious bird species) recovered from a mid-den deposit at Skútustaðir, which lies onthe southern edge of Lake Mývatn, was an-alyzed for δ13C, δ15N, and δ34S, to deter-mine whether the addition of δ34S, to themore traditional suite of δ13C and δ15N,would enable the separation of animals re-ceiving their carbon from terrestrial, ma-rine, or freshwater reservoirs (Sayle et al.2013). The results demonstrated how δ34S,in this ecosystem, allowed for clear differ-entiation between marine, terrestrial, andfreshwater animals. The difference betweenδ34S was considerable between marine andfreshwater animals, with marine values of+15.9 ±1.5‰ and freshwater values of−2.7 ±1.4‰ (Figure 2).

This new information, it was surmised,would be invaluable in better understand-ing human diet, and so these data were usedto interpret the diet of people buried withina cemetery associated with an early chapelat Hofstaðir, which lay


Figure 3. Aerial view of the Hofstaðir site (a), showing the current farm (top), remains of the exca-vated Viking hall (right), and chapel and cemetery under excavation (left). A majority ofthe skeletons were extremely well preserved (b) (images courtesy of Hildur Gestsdóttir).

respectively) (Gestsdóttir 2004). Due tothe very short early settlement period, itis unclear whether the first church pre- orpost-dates the abandonment of a nearbyfeasting hall around AD 1030; however, itis thought that the cemetery was in usebetween the tenth and thirteenth centuries,with all the burials predating the H-1300tephra deposit (AD 1300). In addition, thestratigraphy at the site indicates that the in-tensity of burial in the cemetery was muchgreater in the earlier phases (Gestsdóttir2006; Gestsdóttir and Isaksen 2011).

The ‘raw’ radiocarbon dates from nineindividuals in the cemetery demonstratedthe possibility for a large reservoir off-set (Sayle et al. 2014). Three of the nine(SK016, SK061, and SK066) had radiocar-bon dates that, when calibrated using theterrestrial curve of Reimer et al. (2013)placed their deaths pre-landnám (Table1: Calibrated date). δ13C values from thesenine individuals were used to estimatethe percentage amount of terrestrial andnon-terrestrial protein, and thus degree ofmarine reservoir effect on the radiocarbonmeasurement. Standard linear interpolationmethods were used (Cook et al. 2015) to

determine the percentage diet for eachtype, but the fact that dietary protein formost individuals was coming from ter-restrial, marine, and freshwater systemsmade this methodology both cumbersomeand unsatisfactory. Individuals with a highpercentage of freshwater protein in theirdiet could not be ‘corrected’ for the FRE,given the data available at the time, andtheir dates could only be noted as provid-ing a terminus post quem for their death(Table 1: Re-calibrated date).

In 2014, a further 37 individuals fromthe Hofstaðir cemetery that were beingradiocarbon dated had δ13C, δ15N, and δ34Smeasurements made on their bone collagen(Sayle et al. 2016a). At this time, terrestrialanimal bone (n = 39) from Hofstaðir wasalso subjected to stable isotope analysis. De-spite

Tab

le1

.R

adio

carb

on

ages

and

vari

ou

sm

od

eled

dat

esas

dis

cuss

edin

the

tex

tan

dp

rovi

ded

ingr

eate

rd

etai

lin

Say

leet

al.(2

01

3,

20

14

,2

01

6a,

and

20

16

b).

Th

eM

od

eled

da

teis

the

resu

ltfr

om

Bay

esia

nm

od

elin

gaf

ter

corr

ecti

ng

for

the

FR

Ean

dM

RE

usi

ng

per

cen

tage

sfo

rF

resh

wat

er(%

F)

and

Mar

ine

(%M

)d

iet,

asd

eriv

edfr

om

the

FR

UIT

Sco

mp

ute

rp

rogr

am.

Th

ese

dat

aco

rres

po

nd

toM

od

el2

inSa

yle

etal

.(2

01

6b

)w

her

ea

3.5‰

tro

ph

icsh

ift

was

app

lied

ton

itro

gen

.

Skel

eto

nID

Lab

ID

14C

age

(yr

BP

)C

alib

rate

dd

ate

%M

(±1

0%

)

Re-

cali

bra

ted

dat

e(9

5%

pro

bab

ilit

y)

%F

FR

Eco

rrec

ted

14C

age

(yr

BP

)%

M

Mo

del

edd

ate

(95

%p

rob

-a

bil

ity)

SK00

7SU

ERC

-439

9412

12±

29ca

lAD

690–

890

20ca

lAD

770–

1020

4±

310

48±

139

29±

8ca

lA

D1

06

0–1

24

0

SK00

9SU

ERC

-399

4710

60±

30ca

lAD

890–

1030

27ca

lAD

1010

–123

02

±2

948

±10

641

±7

calA

D1

06

5–1

25

0

SK01

3SU

ERC

-419

7511

23±

24ca

lAD

780–

990

8ca

lAD

770–

1040

5±

488

9±

177

9±

6ca

lA

D1

06

0–1

24

5

SK01

6SU

ERC

-399

5220

30±

3016

0ca

lBC

–cal

AD

6046

calA

D70

–340

∗25

±7

921

±32

117

±8

calA

D1

06

0–1

24

5

SK04

7SU

ERC

-419

8210

05±

24ca

lAD

980–

1150

7ca

lAD

970–

1190

5±

476

9±

181

11±

6ca

lA

D1

06

0–1

25

0

SK05

3SU

ERC

-399

5511

30±

30ca

lAD

770–

990

8ca

lAD

770–

1030

5±

488

1±

185

9±

6ca

lA

D1

06

0–1

24

5

SK05

6SU

ERC

-441

2211

84±

29ca

lAD

720–

950

19ca

lAD

770–

1040

2±

210

72±

106

32±

7ca

lA

D1

06

0–1

24

0

SK06

1SU

ERC

-399

5615

60±

30ca

lAD

420–

570

23ca

lAD

430–

680∗

10±

611

01±

269

18±

8ca

lA

D1

06

0–1

24

5

SK06

6SU

ERC

-399

5717

05±

30ca

lAD

250–

410

36ca

lAD

400–

630∗

14±

610

68±

295

24±

9ca

lA

D1

06

0–1

24

5

∗ TP

Qs.

76 VOLUME 14 • ISSUE 1 • 2019



potential source for interpretative error,these new results cast some doubt on thatassumption and resulted in much of theprevious work being re-evaluated in light ofthe new evidence.

In any robust isotopic study, the fi-nal baseline should focus on temporally re-stricted data provided by the archaeologicalmaterial, but as part of this expanded work,it was felt that the stable isotope values forthe vegetation being consumed by the her-bivores needed characterization in order toaccurately calculate the baseline signal. Thiswas only possible using modern analogues.These results suggested the local baselinewas much more complex than previous re-search indicated, and that the δ34S value forthe Mývatn region was higher than previ-ously predicted due to a possible sea-sprayeffect, despite being approximately 50 kmfrom the coast.

Delving into a study of the modernisotopic values for Lake Mývatn also servedto highlight how stable isotopes can varyacross relatively small distances. Lake Mý-vatn is renowned for its bi-annual hatchingof the midge species, Tanytarsus gracilen-tus. These insects are a crucial componentof the lake’s ecosystem and a key foodsource for its aquatic and bird life (Gar-darsson and Einarsson 2004; Gudbergsson2004; Ives et al. 2008). The midges can forma thick swarm in the immediate vicinity ofthe lake (Figure 4), with an annual inputof 1,200 to 2,500 kg of midges per hectarehaving been observed by Gratton et al.(2008:764), who suggested the midges canproduce “a significant fertilization effect.”This phenomenon decreases logarithmi-cally with distance from the shore, suchthat at 5 km from Lake Mývatn, midgedeposition is negligible. This research wenton to suggest that the massive depositionof midges (δ34S: −3.9‰) in the soil in theimmediate vicinity of the lake (e.g., the areaof Skútustaðir) could potentially lower thesulphur value here, relative to Hofstaðir,while also serving to enrich the δ15N val-ues through fertilization, thus serving tostrengthen the conclusions of Gratton et al.(2008).

Finally, at this point in the research,results indicated several terrestrial herbi-vores with higher bone collagen δ34S val-ues than their contemporaries, suggestingtrade and/or movement of animals to theregion from coastal areas. The broad rang-ing δ13C, δ15N, and δ34S values for humans(−20.2‰ to −17.3‰, 7.4‰ to 12.3‰, and5.5‰ to 14.9‰, respectively) suggestedthe population was consuming varied di-ets, while outliers within the dataset couldconceivably have been migrants to thearea.

Armed with a much more nuanced un-derstanding of the expected stable isotopevalues from the three broad ecosystemsthat combine to form the wider food webfor the Lake Mývatn area (terrestrial, ma-rine, and freshwater), it was decided torevisit the radiocarbon dating of the ceme-tery population in an effort to refine thechronology for the chapel site (Sayle et al.2016b). Here, we moved away from the sim-ple linear interpolation methods that onlyuse one or two stable isotopes, and turnedto new Bayesian statistical tools for produc-ing models for mixed diets. Specifically, thecomputer software program FRUITS (FoodReconstruction Using Isotopic TransferredSignals: Fernandes et al. 2014) was used tocombine the available information on thebaseline values for δ13C, δ15N, and δ34S intoa single model to determine the percentageof terrestrial, marine, and freshwater pro-tein in each 14C-dated individual from thecemetery.

To fully ‘correct’ the radiocarbon age,it was first necessary to produce a realis-tic estimate for the FRE from Lake Mývatn,which required radiocarbon dating modernfish from the lake. The 12 ‘known-age’ Arc-tic charr and brown trout that were datedindicated there was a potentially large FRE.The radiocarbon ages for the modern mate-rial ranged from 3795 to 5329 BP, with anaverage reservoir estimate of 4526 ± 47614C years.

As illustrated in Sayle et al. (2016b), thefreshwater offset was used to first correctthe uncalibrated radiocarbon ages of the46 humans that were presented in the



Figure 4. Image of Lake Mývatn looking from the south near Skútustaðir showing a swarm ofmidges (image courtesy of Árni Einarsson).

previous research. Here, we have repro-duced the results from the nine indi-viduals from the earliest research, withtheir percentage freshwater diet (%F) andFRE-corrected 14C age given in Table 1.Immediately apparent is how the highvariability in the FRE for Lake Mývatn hasgreatly reduced the precision of the orig-inal radiocarbon dates, such that at thispoint there was some skepticism aboutthe utility of the data from this area forproviding an exemplar of the use of theFRUITS program. Nevertheless, the FRE-corrected 14C ages were entered into theOxCal computer program for Bayesianmodeling the chronology for the cemetery.In OxCal, the Mix_Curves function wasused to calibrate the dates by mixing theappropriate percentage of the Marine13calibration curve with the terrestrial calibra-tion curve (IntCal13), both of Reimer et al.(2013). It is these ‘mixed curve’ calibra-tions that were used in subsequent Bayesianmodels.

The two Bayesian models of the studydiffered in the trophic shift applied to nitro-gen (δ15N), with the final results being verysimilar. Model 2, from Sayle et al. (2016b),estimated that burial in the cemetery be-gan in cal AD 1015–1210, lasted for 1–230 years, and ended in cal AD 1125–1300(95% probability; Figure 5). Another pieceof chronological information was availablefrom the cemetery, the tephrochronologicaldating that indicated the cemetery activitymust post-date the V-933 and pre-date theH-1300 tephras. Notably, without applyingany constraints to the start or end bound-aries, all 46 of the modeled dates fell withinthis period (AD 933–1300), thus validatingthe methodology and the results.

SUMMARY

This research highlights a number of po-tential issues when undertaking isotopicstudies in support of a radiocarbon dating

78 VOLUME 14 • ISSUE 1 • 2019


Figure 5. Chronological model for the Hofstaðir cemetery, as modeled in Sayle et al. 2016b. Thefigure shows the nine individuals that have been discussed throughout the paper, whilethe full model is composed of dates from all 46 of the 14C-dated individuals. The verticallines indicate the dates for the V-933 and H-1300 tephras, though these dates were notused as prior information to inform the actual model, thus providing validation of themethodology as applied here.

program, especially in areas where a com-plex food web can lead to multiple carbonreservoirs. It shows a clear need to developisotopic baselines as close to the site of in-terest as possible, and that undertaking anal-yses from across multiple sites of the sametime period in a region has the potentialto uncover nuanced intricacies in the localfood web. Relatively dramatic differenceswere noted across a short distance, and it ispossible that similar differences, not neces-sarily over equally short distances, might beobserved in other ecosystems if these dataare routinely investigated.

In addition to the scale of work thatcan be necessary to develop an isotopicbaseline, this research has revealed justhow important these baselines are for ac-curately calibrating/modeling radiocarbondates. The inclusion of a third stable iso-tope makes these corrections extremely dif-ficult using traditional tools, either linear in-terpolation of multiple pairs or even usingternary diagrams, but the development ofFRUITS has made this less daunting, whilealso making the modeling process transpar-ent and robust.

ACKNOWLEDGEMENTS

This paper is a review of research un-dertaken over a number of years thatwould not have been possible without thesupport of a number of individuals. Theauthors would like to thank the CarnegieTrust for the Universities of Scotland, whopartly funded this research collaborationwith funding for ‘Development of a robustchronology for the archaeology aroundLake Mývatn, Iceland’. We are grateful tothe staff at SUERC for providing materialused in early analyses and assistancewith pretreatment at various stages of theprocess. Particular thanks go to GordonCook for procuring the necessary internalfunding for the pilot study, as well aselements of the follow-up research. HildurGestsdóttir provided samples, valuable ar-chaeological information, and the imagesin Figure 3. Árni Einarsson facilitatedthe first research trip to Iceland, providedinsight into the ecosystem of Lake Mývatn,and Figure 4. A final thanks goes to TomMcGovern and all the team of NABOwho showed enthusiasm for the project,



provided additional samples, and furtherarchaeological details about sites beyondthose found in the published excavationreports.

ORCID

W. Derek Hamilton http://orcid.org/0000-0003-4019-3823Kerry L. Sayle http://orcid.org/0000-0003-3492-5979

REFERENCES

Arnalds, Ó., E. F. Thorarinsdóttir, S. Metusalems-son, A. Johnsson, E. Gretarsson, and A. Arna-son. 1997. Soil Erosion in Iceland. Reykjavik:Icelandic SCS and the Agricultural Research In-stitute.

Ascough, P. L., M. J. Church, G. T. Cook, E.Dunbar, H. Gestsdóttir, T. H. McGovern, A. J.Dugmore, A. Friðriksson, and K. J. Edwards.2012. Radiocarbon reservoir effects in humanbone collagen from northern Iceland. Journalof Archaeological Science 39:2261–2271.

Ascough, P. L., G. T. Cook, M. J. Church, A. J. Dug-more, T. H. McGovern, E. Dunbar, Á. Einars-son, A. Friðriksson, and H. Gestsdóttir. 2007.Reservoirs and radiocarbon: 14C dating prob-lems in Myvatnssveit, northern Iceland. Radio-carbon 49(2):947–961.

Ascough, P. L., G. T. Cook, M. J. Church, E. Dun-bar, Á. Einarsson, T. H. McGovern, A. J. Dug-more, et al. 2010. Temporal and spatial varia-tions in freshwater 14C reservoir effects: LakeMývatn, northern Iceland. Radiocarbon 52(2–3):1098–1112.

Ascough, P. L., G. T. Cook, H. Hastie, E. Dun-bar, M. J. Church, Á. Einarsson, T. H. Mc-Govern, and A. J. Dugmore. 2011. An Ice-landic freshwater radiocarbon reservoir effect:Implications for lacustrine 14C chronologies.Holocene 21(7):1073–1080.

Cook, G. T., P. L. Ascough, C. Bonsall, W.D. Hamilton, N. Russell, K. L. Sayle, E. M.Scott, and J. M. Bownes. 2015. Best prac-tice methodology of 14C calibration of marineand mixed terrestrial/marine samples. Quater-nary Geochronology 27:164–171.

Dugmore, A. J., M. J. Church, P. C. Buckland, K. J.Edwards, I. Lawson, T. H. McGovern, E. Pana-giotakopulu, I. A. Simpson, P. Skidmore, and G.Sveinbjarnardóttir. 2005. The Norse landnám

on the north Atlantic islands: An environmen-tal impact assessment. Polar Record 41:21–37.

Einarsson, Á. 2004. Lake Mývatn and theRiver Laxá: An introduction. Aquatic Ecology38:111–114.

Einarsson, Á., and O. Aldred. 2011. The archaeo-logical landscape of northeast Iceland: A ghostof a Viking Age society. In Remote Sensingfor Archaeological Heritage Management:243–258. Namur: EAC (Europae ArchaeologiaeConsilium) Publications.

Fernandes, R., A. R. Millard, M. Brabec, M. J.Nadeau, P. Grootes. 2014. Food reconstructionusing isotopic transferred Signals (FRUITS): ABayesian model for diet reconstruction. PLOSONE 9(2):e87436.

Gardarsson, A. 2006. Temporal processes andduck populations: Examples from Mývatn. Hy-drobiologia 567:89–100.

Gardarsson, A., and Á. Einarsson. 2004. Resourcelimitation of diving ducks at Mývatn: Food lim-its production. Aquatic Ecology 38:285–295.

Gestsdóttir, H. 2004. Hofstaðir 2003. InterimReport. Fornleifastofnun Íslands. http://www.nabohome.org/uploads/fsi/FS230-910111_Hofstadir_2003.pdf

Gestsdóttir, H. 2006. Hofstaðir 2004. InterimReport. Fornleifastofnun Íslands. http://www.nabohome.org/uploads/fsi/FS311-910112_Hofstadir_2004.pdf

Gestsdóttir, H., and O. Iskasen. 2011. Forn-leifarannsóknirá kirkjugarðiá HofstöðumíMývatnssveitsumarið 2010 (Framvin-duskýrsla). Interim Report. FornleifastofnunÍslands: FS455-910113.

Gratton, C., J. Donaldson, and M. J. Vander Zan-den. 2008. Ecosystem linkages between lakesand the surrounding terrestrial landscape inNortheast Iceland. Ecosystems 11:764–774.

Gudbergsson, G. 2004. Arctic charr in Lake Mý-vatn: The centennial catch record in the lightof recent stock estimates. Aquatic Ecology38:271–284.

Hallsdóttir, M. 1987. Pollen analytical Studiesof Human Influence on Vegetation in Rela-tion to the Landnám Tephra Layer in South-west Iceland. Lundqua Thesis 18, Departmentof Quaternary Geology. Lund: Lund University.

Ives, A. R., Á. Einarsson, VAA Jansen, A. Gardars-son. 2008. High amplitude fluctuations and al-ternative dynamical states of midges in LakeMývatn. Nature 452:84–87.

Jakobsson, S. P., K. Jónasson, and I. A. Sigurds-son. 2008. The three igneous rock series ofIceland. Jökull 58:117–138.

Kristmannsdóttir, H., and H. Ármannsson. 2004.Groundwater in the Lake Mývatn area,

80 VOLUME 14 • ISSUE 1 • 2019

http://orcid.org/0000-0003-4019-3823http://orcid.org/0000-0003-3492-5979http://www.nabohome.org/uploads/fsi/FS230-910111_Hofstadir_2003.pdfhttp://www.nabohome.org/uploads/fsi/FS311-910112_Hofstadir_2004.pdf


northern Iceland: Chemistry, origin and inter-action. Aquatic Ecology 38:115–128.

Lawson, I. T., F. J. Gathorne-Hardy, M. J. Church,A. J. Newton, K. J. Edwards, A. J. Dugmore,and Á. Einarsson. 2007. Environmental im-pacts of the Norse settlement: Palaeoenviron-mental data from Mývatnssveit, northern Ice-land. Boreas 36:1–19.

Lucas, G. 2009. Hofstaðir: Excavations of aViking Age Feasting Hall in North-EasternIceland. Institute of Archaeology, Reykjavík,Monograph Series 1. Reykjavík: Institute of Ar-chaeology.

McGovern, T. H., S. Perdikaris, Á. Einarsson, andJ. Sidell. 2006. Coastal connections, local fish-ing, and sustainable egg harvesting: Patterns ofViking Age inland wild resource use in Mývatndistrict, Northern Iceland. Environmental Ar-chaeology 11:187–205.

McGovern, T. H., O. Vésteinsson, A. Friðriksson,M. Church, I. Lawson, I. A. Simpson, Á.Einarsson, et al. 2007. Landscapes of settle-ment in Northern Iceland: Historical ecologyof human impact and climate fluctuation onthe millennial scale. American Anthropologist109(1):27–51.

Reimer, P. J., E. Bard, A. Bayliss, J. W. Beck, P. G.Blackwell, C. Bronk Ramsey, C. E. Buck, et al.2013. IntCal13 and Marine13 radiocarbon agecalibration curves 0–50,000 years cal BP. Ra-diocarbon 55(4):1869–1887.

Sayle, K. L., G. T. Cook, P. L. Ascough, H. Gests-dóttir, W. D. Hamilton, and T. H. McGovern.2014. Utilization of δ13C, δ15N and δ34S analy-ses to understand 14C-Dating anomalies within

a Late Viking Age community in Northeast Ice-land. Radiocarbon 56(2):811–821.

Sayle, K. L., G. T. Cook, P. L. Ascough,H. R. Hastie, Á. Einarsson, T. H. Mc-Govern, M. T. Hicks, Á. Edwald, and A.Friðriksson. 2013. Application of δ34S anal-ysis for elucidating terrestrial, marine andfreshwater ecosystems: Evidence of animalmovement/husbandry practices in an earlyViking community around Lake Mývatn, Ice-land. Geochimica et Cosmochimica Acta 120:531–544.

Sayle, K. L., W. D. Hamilton, G. T. Cook, P. L.Ascough, H. Gestsdóttir, and T. H. McGov-ern. 2016a. Deciphering diet and monitoringmovement: Multiple stable isotope analysis ofthe Viking Age settlement at Hofstaðir, LakeMývatn, Iceland. American Journal of Phys-ical Anthropology 160:126–136.

Sayle, K. L., W. D. Hamilton, H. Gestsdót-tir, and G. T. Cook. 2016b. Modelling LakeMývatn’s freshwater reservoir effect: Utilisa-tion of the statistical program FRUITS to as-sist in the re-interpretation of radiocarbondates from a cemetery at Hofstaðir, north-east Iceland. Quaternary Geochronology 36:1–11.

Vésteinsson, O. 1998. Patterns of settlement inIceland: A study in prehistory. Saga Book28:1–29.

Vésteinsson, O., T. H. McGovern, and C. Keller.2002. Enduring impacts: Social and environ-mental aspects of Viking age settlement in Ice-land and Greenland. Archaeologia Islandica2:98–136.


ABSTRACTINTRODUCTIONGEOGRAPHICAL AND HISTORICAL BACKGROUNDGEOLOGY OF THE LAKE MÝVATN AREATHE PROBLEMS (AND SOLUTIONS)SUMMARYACKNOWLEDGEMENTSORCIDREFERENCES

stable isotopes, chronology, and bayesian models for the...

Documents