Climate challenges, vulnerabilities, and food security · 2016-01-01 · security, one of the seven human securities identified by a United Nations Human Development Report (15) (see

Climate challenges, vulnerabilities, and food securityMargaret C. Nelsona,b,1, Scott E. Ingramc, Andrew J. Dugmored, Richard Streetere, Matthew A. Peeplesa,Thomas H. McGovernf, Michelle Hegmona, Jette Arneborgg, Keith W. Kintigha, Seth Brewingtonf,Katherine A. Spielmanna, Ian A. Simpsonh, Colleen Strawhackeri, Laura E. L. Comeauj, Andrea Torvinena,Christian K. Madseng, George Hambrechtk, and Konrad Smiarowskif

aSchool of Human Evolution and Social Change, Arizona State University, Tempe, AZ 85287; bBarrett Honors College, Arizona State University, Tempe, AZ85287; cDepartment of Sociology and Anthropology, University of Texas at Arlington, Arlington, TX 76019; dSchool of GeoSciences, University of Edinburgh,Edinburgh EH8 9XP, Scotland, United Kingdom; eDepartment of Geography and Sustainable Development, University of St. Andrews, St. Andrews, FifeKY16 9AL, Scotland, United Kingdom; fAnthropology Department, Hunter College, City University of New York, New York, NY 10065; gDepartment forDanish Middle Ages and Renaissance, National Museum, DK-1220 Copenhagen, Denmark; hBiological and Environmental Sciences, School of NaturalSciences, University of Stirling, Stirling FK9 4LA, Scotland, United Kingdom; iNational Snow and Ice Data Center, University of Colorado Boulder, Boulder, CO80309; jTransport for London, London SW1H 0TL, United Kingdom; and kAnthropology Department, University of Maryland, College Park, MD 20817

Edited by John A. Dearing, University of Southampton, Southampton, United Kingdom, and accepted by the Editorial Board November 24, 2015 (received forreview April 10, 2015)

This paper identifies rare climate challenges in the long-term historyof seven areas, three in the subpolar North Atlantic Islands and fourin the arid-to-semiarid deserts of the US Southwest. For each case,the vulnerability to food shortage before the climate challenge isquantified based on eight variables encompassing both environ-mental and social domains. These data are used to evaluate therelationship between the “weight” of vulnerability before a climatechallenge and the nature of social change and food security follow-ing a challenge. The outcome of this work is directly applicable todebates about disaster management policy.

vulnerability | climate challenge | disaster management | prehistory |history

Managing disasters, especially those that are climate-induced,calls for reducing vulnerabilities as an essential step in re-

ducing impacts (1–8). Exposure to environmental risks is but onecomponent of potential for disasters. Social, political, and eco-nomic processes play substantial roles in determining the scale andkind of impacts of hazards (1, 8–12). “Disasters triggered by nat-ural hazards are not solely influenced by the magnitude and fre-quency of the hazard event (wave height, drought intensity etc.),but are also rather heavily determined by the vulnerability of theaffected society and its natural environment” (ref. 1, p. 2). Thus,disaster planning and relief should address vulnerabilities, ratherthan returning a system to its previous condition following a di-saster event (6).Using archaeologically and historically documented cultural

and climate series from the North Atlantic Islands and the USSouthwest, we contribute strength to the increasing emphasis onvulnerability reduction in disaster management. We ask whetherthere are ways to think about climate uncertainties that can helppeople build resilience to rare, extreme, and potentially devas-tating climate events. More specifically, we ask whether vulnera-bility to food shortfall before a climate challenge predicts the scaleof impact of that challenge. Our goal is both to assess current un-derstandings of disaster management and to aid in understandinghow people can build the capability to increase food security andreduce their vulnerability to climate challenges.We present analyses of cases from substantially different re-

gions and cultural traditions that show strong relationships be-tween levels of vulnerability to food shortage before rare climateevents and the impact of those events. The patterns and detailsof the different contexts support the view that vulnerability cannotbe ignored. These cases offer a long-term view rarely included instudies of disaster management or human and cultural well-being(for exceptions, see refs. 13 and 14). This long time frame allows usto witness changes in the context of vulnerabilities and climatechallenges, responding to a call for more attention to “how humansecurity changes through time, and particularly the dynamics

of vulnerability in the context of multiple processes of change”(ref. 10, p. 17).

ApproachIn this study, we focus on climate challenges that can impact foodsecurity, one of the seven human securities identified by a UnitedNations Human Development Report (15) (see also ref. 10) andone of the core components of human well-being as identified bythe Millennium Ecosystem Assessment Board (16). Food secu-rity refers to “physical and economic access to basic food” (ref.15, p. 27). Integral to our perspective is a multidimensionalconceptualization of food security as involving both the avail-ability of food and access to that food (e.g., 17, 18). The capa-bility of people to access food can be limited by structural andsocial conditions (19, 20), as we identify in this study.We use the concept of vulnerability to assess resilience of food

security to climate challenges. Resilience is the ability of a systemto absorb disturbances without losing its identity (21) and itscapacity to absorb perturbations or shocks while maintainingessential structures and functions (22, 23). Vulnerability is “thestate of susceptibility to harm from exposure to stresses associated

Significance

Climate-induced disasters are impacting human well-being inever-increasing ways. Disaster research and management rec-ognize and emphasize the need to reduce vulnerabilities, al-though extant policy is not in line with this realization. Thispaper assesses the extent to which vulnerability to food short-age, as a result of social, demographic, and resource conditionsat times of climatic challenge, correlates with subsequent de-clines in social and food security. Extreme climate challenges areidentified in the prehispanic US Southwest and historic Norseoccupations of the North Atlantic Islands. Cases with such dif-ferent environmental, climatic, demographic, and cultural andsocial traditions allow us to demonstrate a consistent relation-ship between vulnerability and consequent social and food se-curity conditions, applicable in multiple contexts.

Author contributions: M.C.N., S.E.I., A.J.D., R.S., M.A.P., T.H.M., M.H., J.A., K.W.K., S.B.,and K.A.S. designed research; M.C.N., S.E.I., A.J.D., R.S., M.A.P., T.H.M., M.H., J.A., K.W.K.,S.B., K.A.S., I.A.S., C.S., L.E.L.C., A.T., C.K.M., G.H., and K.S. performed research; M.C.N., S.E.I.,A.J.D., R.S., M.A.P., T.H.M., M.H., J.A., K.W.K., S.B., K.A.S., C.S., and A.T. analyzed data; andM.C.N., S.E.I., A.J.D., R.S., M.A.P., T.H.M., and M.H. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission. J.A.D. is a guest editor invited by theEditorial Board.

Freely available online through the PNAS open access option.1To whom correspondence should be addressed. Email: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1506494113/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1506494113 PNAS Early Edition | 1 of 6

ANTH

ROPO

LOGY

http://crossmark.crossref.org/dialog/?doi=10.1073/pnas.1506494113&domain=pdf&date_stamp=2015-12-23

mailto:[email protected]

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1506494113/-/DCSupplemental

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1506494113/-/DCSupplemental

www.pnas.org/cgi/doi/10.1073/pnas.1506494113

with environmental and social change and from the absence ofcapacity to adapt” (ref. 24, p. 268). Turner and colleagues (9)identify exposure, sensitivity, and resilience as key components ofvulnerability. Our study focuses specifically on Turner et al.’s di-mension of sensitivity. We examine conditions that impact thecapability of people to maintain food security, including bothavailability and access. Vulnerability to climate challenges is me-diated by institutional structures (23) (see also refs. 11 and 25) thatare constantly changing and impacting people’s capabilities toavoid declines in food security.Disaster managers are especially concerned with vulnerabil-

ities, the preconditions that lead climate challenges such asdroughts, floods, and extreme cold conditions to become disas-ters, recognizing that it is at the interface of environmental andsocial conditions that disasters occur (9, 12, 13, 26). Our researchbuilds on arguments that resilience to the impacts of climate(and other) challenges can be built by reducing vulnerabilities(2–6, 9, 12). However, people “tend to push the risk spectrumtoward catastrophic events occurring with increasing probability”(ref. 14, p. 8).To explore the relationship between vulnerability, food secu-

rity, and the impacts of climate challenges, we quantify social andclimate conditions in seven centuries-long sequences. First, weidentify 13 points in our climate sequences that are rare andextreme. We then quantify the extent of vulnerability to foodshortfall for the period immediately preceding each climateevent. Finally, we identify the conditions following each climateevent in terms of major social changes and declines in food se-curity, specifically food shortage. We compare these conditionswith the vulnerability before each climate challenge to considerthe role of vulnerabilities in the impact of climate challenges.

The CasesArchaeological and historical cases are used to examine the roleof vulnerability in climate impacts. Two features of the cases areparticularly important. First, each is a long record of coupledsocial and environmental change, with data on demography,social institutions and traditions, food economies, political rela-tions, and climate conditions. This long-term record documentsthe contexts and impacts of climate challenges. Building ro-bustness to climate challenges is a daunting task complicated bylimitations of current and recent experience on scenarios ofpossible challenges and solutions (14, 27). Long historical se-quences provide series of known changes in human–landscape–climate interactions that represent a set of completed experimentsin human ecodynamics (12, 28–31). We use these sequences toidentify when rare climate events occurred, what the vulnerabilityload was before each event, and the scale and type of changesfollowing each climate challenge. This requires a window in timemuch longer than is usually available from contemporary experi-ences (see also refs. 14 and 27), although local and traditionalknowledge offers some perspective on vulnerability to long-termor rare processes (32).Second, the cases are from very different regions of the world—

the arid, warm deserts of the US Southwest and the subarcticregion of the North Atlantic Islands. Patterns in one region orimpacts of one type of climate challenge may be informative onlyfor that region. The cases we compare are from different climateregimes, physiographic regions, cultural traditions, and historicalcontexts. Patterns evident in this diverse database indicate rela-tionships between vulnerability and the impacts of extreme climateevents that have import for resilience planning and disastermanagement generally.

North Atlantic. The North Atlantic cases include Norse occupa-tions in Iceland (33, 34), Greenland (35), and the Faroe Islands(36) beginning in the late 9th to late 10th centuries and extendinginto the 18th century, except in Greenland, depopulated by the

Norse in the 15th century. Data derive from decades of climate,historical, and archaeological research by the North AtlanticBiocultural Organisation (NABO; www.nabohome.org) (SI Ap-pendix, section 2), which promotes international and interdisci-plinary research collaboration. Recurring research themes havebeen colonization and interactions of human–environmental impact,climate change, and early globalization that have produced re-markably different outcomes on the millennial scale (e.g., 34, 37–40).

US Southwest. The cases from the US Southwest are all in-digenous occupations of what is now Arizona and New Mexicoduring the 10th to 16th centuries. Data derive from researchteams within the Long-Term Vulnerability and TransformationProject (LTVTP; ltvtp.shesc.asu.edu) that conduct field researchin the Zuni (41, 42), Salinas (43), Mimbres (44, 45), and Ho-hokam (46, 47) archaeological regions. LTVTP researchers ex-amine the relationships between vulnerabilities in the social andecological realms and the magnitude and scale of social–eco-logical transformations (48), comparing long sequences ofchange and stability. These sequences illustrate the extent towhich short-term strategies create vulnerabilities that play outover time.These archaeological and historical sequences are not sources

of “lessons” as much as they are sources of information on howdecisions and actions created vulnerabilities and how these vul-nerabilities played out over time under different challenges [seealso Turner and Sabloff (13) for the Classic Maya; Tainter (27)for problem solving and collapse in the Roman Empire; andButzer (26) for a collection of historical studies]. This researchposits that existing vulnerabilities to food shortages can be trig-gered by rare climate challenges, for which planning and antic-ipation are difficult. Planning that includes a focus on keepingvulnerabilities low can contribute to resilience to unanticipated(or unpredictable) climate challenges (9).

Results and DiscussionRare Climate Challenges. Across seven regions, four in the pre-hispanic US Southwest and three that are Norse occupations ofthe North Atlantic Islands, various kinds of climatic records areused to identify rare climate challenges with considerable po-tential to result in “disaster.” For the US Southwest, we identi-fied dry periods (droughts) in annual tree-ring proxy records ofprecipitation and streamflow (SI Appendix, section 1). Dry pe-riods decreased the productivity of the resources people reliedon for food. Climate conditions associated with each case arerepresented by separate climate reconstructions that begin be-tween 436 and 879 C.E. The rare climate challenges identified inTable 1 (fourth column) are the longest (15–23 y in duration)and rarest (they had not occurred for at least 456 y) dry periodsof the 10th through 16th centuries, and most are the longest ineach reconstruction. Although dry periods were common in theregion, the challenges to the food security of farmers were likelyunprecedented during these long and rare dry periods. For theNorth Atlantic, challenges are rare extremes or regime changesfor climate systems that involve cold temperatures, sea ice, and/or storminess (Table 1, fourth column). Proxy records of tem-perature, sea ice, and storminess are used to identify climatechallenges during the period 900–1900 C.E. (SI Appendix, section 1).These proxy records have strong spatial coverage but relativelypoor chronological resolution relative to the US Southwest. Ex-treme events were identified by both large (at least one sigma)deviation from the previously experienced long-term mean anduniqueness—the event was not experienced in the previous 200 y.Climate regime change (e.g., the onset of the so-called Little IceAge) events were prioritized if they were the first experienceddeviation from the previous normal, even if subsequent largerdeviations occurred. Events had to be recognized in two proxy

2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1506494113 Nelson et al.

http://www.nabohome.org/

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1506494113/-/DCSupplemental/pnas.1506494113.sapp.pdf


http://ltvtp.shesc.asu.edu




records to confirm a climate challenge. Some events were un-precedented in Norse experience on the North Atlantic Islands.

Vulnerability Loads. How vulnerable were people to shortfall infood supply, given the configuration of social and environmentalconditions, before each identified climate challenge?We quantify the “load” of vulnerability to food shortage be-

fore these climate challenges using eight variables grouped intotwo domains: (i) population–resource, which has to do primarilywith the overall availability of food relative to population size;and (ii) social institutions and practices, which have more to dowith access to food including through social and economicstructures (Table 2). With this characterization, we identify thekinds of conditions contributing to vulnerability and the overallload of vulnerability for each case. By load, we mean the extentto which each variable contributed to the likelihood that peoplemight experience impacts from climate challenges (for a related

approach, see ref. 50). We used a qualitative ranking of the stateof each variable to quantify its contribution to the vulnerabilityload (Table 1, Right). The rankings ranged from no contribution(1) to substantial contribution (4) to vulnerability. Codes of 2and 3 capture conditions that were minor (2) to more substantial(3) but not as strong as the ends of the continuum. Coding wasbased on expert knowledge of case leaders using evidence fromarchaeological and historical records (SI Appendix, section 2).The vulnerability load for a case is represented by the “total” ofthese scores (Table 1, Far Right); a score of 8 indicates no vul-nerability presented by any of the variables, whereas a score of 32indicates strong contributions from all variables to the totalvulnerability load.Differences in the mean contribution of the variables to the

vulnerability load illustrate the importance of social institutions andissues of access to food in managing vulnerability (Table 1, Bottom).

Table 1. Rare climate challenges, vulnerability scores, and total vulnerability load

Vulnerability scores for each variable

Region Case Initiation date, C.E. Kind of challenge V1 V2 V3 V4 V5 V6 V7 V8 T

US SW Z 1133 Extreme dry 1 2 2 2 2 1 1 1 12S 1335 Extreme dry 1 2 1 3 2 4 1 3 17

M 1 1127 Extreme dry 2 2 3 4 2 2 2 1 18M 2 1273 Extreme dry 1 2 1 1 2 1 1 1 10H 1 1338 Extreme dry 2 2 2 4 2 4 4 3 23H 2 1436 Extreme dry 3 2 3 4 2 4 4 3 25

NA G 1 1257 Extreme cold 1 2 2 2 1 3 3 2 16G 2 ca. 1310 RC: colder system, increasing

sea ice1 3 2 3 1 4 3 2 19

G 3 ca. 1421 RC: stormier, extreme cold 1 4 3 4 2 4 1 3 22I 1 1257 Extreme cold 1 1 2 1 1 2 2 2 12I 2 ca. 1310 RC: colder system with

sea ice2 1 2 1 1 2 2 2 13

I 3 1640 Extreme cold, sea icegreatest extent

2 1 3 2 2 2 2 2 16

F 1257 Extreme cold 2 1 2 2 2 3 2 2 16Mean vulnerability score for each variable 1.5 1.9 2.2 2.5 1.7 2.8 2.2 2.1

F, Faroes; G, Greenland; H, Hohokam; I, Iceland; M, Mimbres; NA, North Atlantic; RC, climate regime change;S, Salinas; T, total vulnerability load; US SW, US Southwest; V1, available food; V2, resource diversity; V3, resourcedepression; V4, connection; V5, storage; V6, mobility; V7, equal access; V8, barriers; Z, Zuni.

Table 2. Variables contributing to vulnerability load to food shortage

Vulnerability variables Evidence for vulnerability Value of variable for resilient food system

Population–resourceconditionsAvailability of food Insufficient calories or nutrients Balance of available resources and population

size reduces risk of shortfallDiversity of available,

accessible foodInadequate range of resources responsive

to varied conditionsDiverse portfolio reduces risk, increases options (9)

Health of food resources Depleted or degraded resources, habitats Healthy habitats contribute to managingrisk and change (26, 49)

Social conditionsConnections Limited connections with others experiencing

different conditionsSocial networks expand access to food and land (26)

and are sources for risk pooling (49)Storage Insufficient, inaccessible storage Stored foods reduce risk in times of shortageMobility Inability to move away from challenging

food conditionsMovement to alternative places, landscapes,

and social groups offers potential for addressingresource shortfall through access to food/land (49)

Equal access Unequal control and distribution of land, water,and food resources

Equal access avoids challenges to coping and adaptivecapacity in disaster risk management

Barriers to resource areas Physical barriers limiting access tokey resource areas

Lack of barriers enhances capability of people to provisionthemselves with food

Nelson et al. PNAS Early Edition | 3 of 6

ANTH

ROPO

LOGY


For the full set of 13 cases, two social domain variables—connec-tions and mobility—contribute more to the vulnerability load thando any other variables, as indicated in the mean scores shown acrossthe bottom of the table. In contrast, lack of an adequate food supply(V1) rarely contributed much to vulnerability.This pattern is consistent with issues identified by disaster

managers, who emphasize that social factors and insufficienciesin aid-related resources limit their abilities to reduce vulnera-bilities before extreme climate events (3, 6). Governments andnongovernmental organizations (NGOs) are often loath to al-locate or raise funds to change conditions when the generalpopulation is not actually experiencing food shortfalls. As a re-sult, disaster management is often oriented toward recovery andresponse to crises that may have been avoided or reduced hadprior vulnerabilities been addressed.

Social Change Following Extreme Climate Challenges. Were climatechallenges followed by major social change? Fig. 1 (Left) showsthe relationship between vulnerability load before a climate eventand the extent of social change following that event. The distri-bution of vulnerability loads for the cases is plotted in ascendingorder, with colors indicating whether social changes followed theshock. “Transformation” (red) refers to circumstances of bothconsiderable population decline and disappearance of key socialinstitutions and structures (51). “Substantial change” (orange) in-dicates changes in social institutions and structures without de-mographic decline. These rankings are based on evidence of changein household and village form and count, change in communitystructures, and historical records describing the scale and magni-tude of change (SI Appendix, section 3).Little change occurred only at low levels of vulnerability load;

transformation occurred where vulnerability loads were quitehigh. Substantial change occurred across the spectrum of vulner-ability loads.The end of the Norse occupation in Greenland (Greenland 3),

the population decline and end of a cultural tradition in Mimbres

(Mimbres 1), and the depopulation and institutional collapse inthe Hohokam area of central Arizona (Hohokam 2) exemplifytransformation. In Greenland, the eastern settlement was aban-doned by the Norse around 1450 (35). The challenges to people inGreenland were many, including radical decreases in food securityand isolation from their original northern European homeland.Norse settlement of Greenland ended shortly after the shock welist as Greenland 3, either because the last settlers died or peoplefound their way off the island at that time (52).In the Mimbres case, nearly everyone left their village settle-

ments during a severe and long drought event that began in 1127.People had depleted riverine habitats in some parts of the region(53), decreased the abundance of artiodactyls (54, 55), anddamaged some upland soils through farming (56). In addition,external relations with other Southwestern groups appear tohave been severely limited (57). Nelson and colleagues (30) haveestimated that roughly three-quarters of the population (ref. 30,figure 4) migrated away, and Mimbres traditions evident in manymaterial domains ceased.The Hohokam canal irrigation systems in central Arizona were the

largest in pre-Columbian North America. By the mid-1400s, thou-sands of people had emigrated from these systems and large asso-ciated settlement clusters, leaving little visible archaeological trace ofvillages or settlements (51). Some may have died from malnutritionat some villages (46, 58) but most moved away, leaving an all butunpopulated center that once had supported many thousands.

Food Shortage Following Extreme Climate Challenges. Was eachclimate event followed by food shortage? Fig. 1 (Right) shows therelationship between vulnerability to food shortage before an eventand the experience of food shortage following each event. Coding ofthe experience of food shortage was based on evidence fromskeletal analysis of humans and animals, paleoethnobotanicalanalysis, and historical documents (SI Appendix, section 3). “None”(green) indicates no evidence of shortage; “some” (orange) indi-cates some food shortage for all people or substantial shortage for

Fig. 1. Social changes and food conditions following climate challenges.





some people; and “substantial” (red) marks those cases with sub-stantial shortage for all.No shortage (green) following climate events is evident in six

cases. Among those cases, Mimbres 1 and Greenland 2 offerperspectives on disaster management. In the Mimbres 1 case,most of the regional population emigrated, which reduced thelocal population to a level that avoided food provisioning issuesbut which had dramatic impacts socially—the transformativechange noted above. In the Greenland case, people shifted to-ward substantial reliance on marine mammals (59–61), narrow-ing their diet by focusing on a resource that was abundant at thattime. However, this narrowing of diet increased vulnerability toshortfall, which was realized just over a century later when theNorse occupation of Greenland ended just after the climatechallenge we label as Greenland 3 (35, 62). In both cases, foodshortage was avoided but at a high cost.In 7 of the 13 cases, food shortage is evident at some level

(yellow or red). The three Iceland cases at the lower end of thevulnerability load spectrum were contexts of persistent hunger.Climate challenges increased the extent of hunger but never toextreme levels for the whole population. Vulnerability to foodshortfall remained low throughout, perhaps because people wereaware and responsive to the reality that hunger was a constantchallenge. These low vulnerability loads may have played a rolein preventing extreme shortages following climate challenges.Streeter and colleagues (63) have noted that Icelandic societywas consistently resilient to an array of challenges, bouncing backfrom plagues, conflicts, and difficult climate conditions.The three cases with the highest vulnerability loads (Fig. 1,

Lower Right) all have evidence of food shortage. Hohokam 1 is aperiod, beginning in 1338, when there is some evidence of foodshortage for one segment of the population in central Arizona(46). By the second dry period, Hohokam 2, beginning in 1436,nearly everyone had left the massive irrigation systems. We in-terpret this as evidence of substantial food shortage, because itresulted in the disuse of a massive irrigation system and largeamounts of previously cultivated land. The Greenland shock thatbegan in 1421 coincides with the end of the Norse occupation inGreenland, which has been attributed to a variety of challengingconditions of which access to the key food resource—off-shoreseals—is but one (62).

Summary and RecommendationsThis analysis of historically and archaeologically documented casesfrom substantially different regions and cultural traditions shows aconsistent relationship between the load of vulnerability to foodshortage before a challenging climate event and the scale of impactfollowing that challenge. Major social changes and food shortfallfollowed climate challenges in the cases with the highest existingvulnerability loads. Social change and food shortage were less oftenexperienced and were never extreme in the cases with lowest vul-nerability. The pattern is consistent across different regions of theworld experiencing substantially different climate conditions—therole of vulnerability cannot be ignored.Our stated goal was to assess current understandings of disaster

management and to aid in understanding how people can buildcapability to increase food security and reduce their vulnerabilityto climate challenges. Our analysis suggests several points in thisregard that are well-understood in the risk management commu-nity even though changes to vulnerability remain elusive and di-sasters grow more common (2, 3, 5).

i) Strategies for coping with climate challenges should includefocus on the reduction of vulnerabilities, which disaster man-agers and others identify as an essential step in reducing im-pacts (1–8). The climate events we document were trulyunanticipated, yet in those cases with low vulnerability loadswe find little or no evidence for major impacts. What are often

called “natural disasters” were avoided by maintaining condi-tions, especially social conditions, that kept vulnerability low (9).

ii) Supporting the work of many others (1, 8–12), our analysisdemonstrates that social factors are substantial contributorsto vulnerability. Although researchers and managers recog-nize the role of social conditions, management of food secu-rity may address simply the availability of food resulting frompopulation–resource balance. We can err in our managementof food security by assuming that in contexts of adequate foodavailability there is no vulnerability to food shortage. Atten-tion to social conditions that create vulnerabilities to foodshortage is essential in resilience to climate challenges.

iii) As many have noted (e.g., 8, 9, 11–14), disasters are notinevitable; they are the result, in large part, of human-madeconditions. The concept of natural disaster is unfortunatebecause it removes focus from the social conditions thatset the stage for disasters to be triggered by various chal-lenges. Our diverse cases suggest that human-created vul-nerability can influence the outcome of climate challenges inmany environmental, cultural, and historical contexts.

iv) Disaster relief should include addressing vulnerabilities,rather than returning systems to previous conditions (6). Werecognize that change from untenable conditions is difficult(64). However, Kinver (3), reporting on responses to famine,notes consistent evidence that early action is cheaper.

Debates about disaster management, responses to climateshocks, attainment of human securities, and resilience to un-certainty rarely benefit from long time spans over which toevaluate claims. Our analyses offer a long-term view that allowsassessment of full cycles of coupled social–ecological systems.Our work demonstrates that at the lowest and highest levels ofvulnerability load, impacts are felt from climate challenges indifferent climate and social contexts. The pattern of outcomesfrom this cross-case study of different cultures, traditions, times,and environments underscores the critical need for reducingvulnerabilities to food shortfall to avoid the actual experience ofshortage and painful social changes. And this, we hope, can helpmove discussions and actions forward.

Materials and MethodsClimate reconstructions and identification of climate challenges used an arrayof data sources and techniques (SI Appendix, section 1). The climate chal-lenges identified in the US Southwest are the longest and rarest dry periodsduring the focal period. To represent climatic conditions for each case, weselected annually resolved tree-ring precipitation and streamflow recon-structions closest to the primary settlement areas of each case. Each re-construction was smoothed with a centered 9-y-interval moving average toidentify trends in the data obscured by year-to-year variation inherent inmost arid regions. Years in the first quartile of the distribution of intervalaverages of each reconstruction were classified as dry periods. For eachidentified dry period, we calculated the number of years since a dry periodof equal or greater duration had occurred. The dry periods classified as rareclimate challenges had not been experienced by people or the arid envi-ronment they relied on for at least 456 y.

The climate challenges identified in the North Atlantic include coldsummer temperatures, increased sea ice occurrence, and increased stormi-ness. These were identified from ice cores, marine cores, lake sediments, andglaciological records. Extremes are identified as extended (>3-y) deviationsgreater than 1 SD from the mean from ice core and multiproxy climate re-constructions. Climate reconstructions from global circulation models wereused to identify additional large-magnitude cooling events due to volcanicforcing. Due to the relatively poor chronological and spatial resolution ofthese datasets, we only considered a climate event significant if it was ob-served in more than one record. Although the chronology is not as precise aswe would like for the regime shift scenarios, change would have been rapidin both human and environmental terms. The multiyear climate extremesidentified in both regions decreased the productivity of resources peoplerelied on and increased the risk of food shortages. We can infer fromour current knowledge of artic and subarctic ecological systems (e.g., theimpact of summer temperatures on reducing growing-season length) and

Nelson et al. PNAS Early Edition | 5 of 6

ANTH

ROPO

LOGY


documentary records [e.g., high rates of livestock mortality on Inuit farmsassociated with cold weather (65)] that these climate events would have hadsignificant consequences and could be considered “shocks.”

Calculation of vulnerability load used historical, archaeological, andpaleoethnobotanical data (SI Appendix, section 2). We used a qualitativeranking of the state of each variable to quantify the contribution of each tothe “vulnerability load.” These rankings represent expert knowledge basedon empirical evidence for each case (SI Appendix, section 2). Experts wereassembled and participated together in coding.

Conditions following challenges were coded from empirical historical andarchaeological data for social change and change in access to food (SI Ap-pendix, section 3).

ACKNOWLEDGMENTS. This research is funded by National Science Founda-tion (NSF) Dynamics of Coupled Natural and Human Systems BCS-1113991;NSF Arctic Social Science Research Coordination Network (RCN) ARC-1104372; and NSF Polar Programs Science, Engineering, and EducationRCN 1140106. The Amerind Foundation and the Santa Fe Institute hostedworking group meetings for this collaboration.

1. Birkmann J (2007) Climate change & vulnerability: Challenges for spatial planning andcivil protection. Forum DKKV/CEDIM: Disaster Reduction in Climate Change (KarlsruheUniversity, Karlsruhe, Germany).

2. Few R, Osbahr H, Bouwer LM, Viner D, Sperling F (2006) Linking Climate ChangeAdaptation and Disaster Risk Management for Sustainable Poverty Reduction:Synthesis Report (VARG, Washington, DC).

3. Kinver M (April 5, 2013) Chatham House report: Famine risks are badly managed. BBCNews Science & Environment.

4. Matthew RA, Barnett J, McDonald B, O’Brien KL, eds (2010) Global EnvironmentalChange and Human Security (MIT Press, Cambridge, MA).

5. O’Brien G, O’Keefe P, Rose J, Wisner B (2006) Climate change and disaster manage-ment. Disasters 30(1):64–80.

6. O’Brien K, et al. (2008) Disaster risk reduction, climate change adaptation and humansecurity. GECHS Report 2008:3.

7. Thomalla F, Downing T, Spanger-Siegfried E, Han G, Rockström J (2006) Reducinghazard vulnerability: Towards a common approach between disaster risk reductionand climate adaptation. Disasters 30(1):39–48.

8. Turner BL, II, et al. (2003) Illustrating the coupled human-environment system forvulnerability analysis: Three case studies. Proc Natl Acad Sci USA 100(14):8080–8085.

9. Turner BL, II, et al. (2003) A framework for vulnerability analysis in sustainabilityscience. Proc Natl Acad Sci USA 100(14):8074–8079.

10. Barnett J, Matthew RA, O’Brien KL (2010) Global environmental change and humansecurity: An introduction. Global Environmental Change and Human Security, edsMatthew RA, Barnett J, McDonald B, O’Brien KL (MIT Press, Cambridge, MA), pp 3–32.

11. Hewitt K (1983) The idea of calamity in a technocratic age. Interpretations of Calamity:From the Viewpoint of Human Ecology, ed Hewitt K (Allen & Unwin, Boston), pp 3–32.

12. Hewitt K (1997) Regions of Risk: A Geographical Introduction to Disasters (Longman,Essex, UK).

13. Turner BL, II, Sabloff JA (2012) Classic Period collapse of the Central Maya Lowlands:Insights about human-environment relationships for sustainability. Proc Natl Acad SciUSA 109(35):13908–13914.

14. Dearing JA, Braimoh AK, Reenberg A, Turner BL, van der Leeuw S (2010) Complex landsystems: The need for long time perspectives to assess their future. Ecol Soc 15(4):21.

15. United Nations Development Program (1994) Human Development Report 1994(Oxford Univ Press, New York).

16. Millennium Ecosystem Assessment Board (2005) Living Beyond Our Means: NaturalAssets and Human Well-Being. Available at www.millenniumassessment.org.

17. Barrett CB (2010) Measuring food insecurity. Science 327(5967):825–828.18. FAO, IFAD, and WFP (2014) The State of Food Insecurity in the World 2014: Strength-

ening the Enabling Environment for Food Security and Nutrition (FAO, Rome).19. Sen A (1999) Development as Freedom (Anchor, New York).20. Sen A (2005) Conceptualizing and measuring poverty. Poverty and Inequality, eds

Grusky DB, Kanbur R (Stanford Univ Press, Palo Alto, CA), pp 30–46.21. Folke C (2006) Resilience: The emergence of a perspective for social-ecological systems

analyses. Glob Environ Change 16(3):253–267.22. Holling CS (1973) Resilience and stability of ecological systems. Annu Rev Ecol Syst 4:1–23.23. Holling CS, Gunderson LH (2002) Resilience and adaptive cycles. Panarchy:

Understanding Transformations in Human and Natural Systems, eds Gunderson LH,Holling CS (Island, Washington, DC), pp 25–62.

24. Adger WN (2006) Vulnerability. Glob Environ Change 16(3):268–281.25. Burton I, Kates RW, White GF (1993) The Environment as Hazard (Guilford, New York).26. Butzer KW (2012) Collapse, environment, and society. Proc Natl Acad Sci USA 109(10):

3632–3639.27. Tainter JA (2000) Problem solving: Complexity, history, sustainability. Popul Environ 22(1):3–41.28. Costanza R, et al. (2007) Sustainability or collapse: What can we learn from inte-

grating the history of humans and the rest of nature? Ambio 36(7):522–527.29. Dearing JA (2011) Learning from the past. Global Change 77:16–19.30. Nelson MC, Kintigh KW, Abbott DR, Anderies JM (2010) The cross-scale interplay

between social and biophysical context and the vulnerability of irrigation-dependentsocieties: Archaeology’s long-term perspective. Ecol Soc 15(3):31.

31. van der Leeuw S, et al. (2011) Toward an integrated history to guide the future. EcolSoc 16(4):2.

32. Berkes F, Colding J, Folke C (2000) Rediscovery of traditional ecological knowledge asadaptive management. Ecol Appl 10(5):1251–1262.

33. Vésteinsson O, McGovern TH (2012) The peopling of Iceland. Nor Archaeol Rev 45(2):206–218.

34. McGovern TH, et al. (2007) Landscapes of settlement in northern Iceland: Historicalecology of human impact and climate fluctuation on the millennial scale. AmAnthropol 109(1):27–51.

35. Arneborg J (2008) The Norse settlements in Greenland. The Viking World, eds Brink S,Price N (Routledge, New York), pp 588–597.

36. Arge SV, Sveinbjarnardóttir G, Edwards KJ, Buckland PC (2005) Viking and Medievalsettlement in the Faroes: People, place, and environment. Hum Ecol 33(5):597–620.

37. Adderley WP, Simpson IA, Vésteinsson O (2008) Local scale adaptations: A modeledassessment of soil, landscape, microclimatic and management factors in Norse home-field productivities. Geoarchaeology 23(4):500–527.

38. Dugmore AJ, Keller C, McGovern TH (2007) Norse Greenland settlement: Reflectionson climate change, trade, and the contrasting fates of human settlements in theNorth Atlantic Islands. Arctic Anthropol 44(1):12–36.

39. Simpson IA, et al. (2002) Soil limitations to agrarian land production in premodernIceland. Hum Ecol 30(4):423–443.

40. Thomson AM, Simpson IA, Brown JL (2005) Sustainable rangeland grazing in NorseFaroe. Hum Ecol 33(5):737–761.

41. Kintigh KW, Glowacki DM, Huntley D (2004) Long-term settlement history and theemergence of towns in the Zuni area. Am Antiq 68(4):432–456.

42. Peeples MA, Schachner G, Kintigh KW The Cibola/Zuni Region. The Oxford Handbookof Southwest Archaeology, eds Mills BJ, Fowles S (Oxford Univ Press, Oxford).

43. Spielmann KA, ed (2011) The Salinas province: Archaeology at the edge of the Puebloworld. Archaeol Southwest 25(2).

44. Hegmon M (2002) Recent issues in the archaeology of the Mimbres region of theNorth American Southwest. J Archaeol Res 10(4):307–357.

45. Nelson MC, Hegmon M, Kulow S, Schollmeyer KG (2006) Archaeological and eco-logical perspectives on reorganization: A case study from the Mimbres region of theUS Southwest. Am Antiq 71(3):403–432.

46. Abbott DR, ed (2003) Centuries of Decline During the Hohokam Classic Period atPueblo Grande (Univ of Arizona Press, Tucson, AZ).

47. Ingram SE (2010) Human vulnerability to climatic dry periods in the prehistoric U.S.Southwest. PhD dissertation (Arizona State University, Tempe, AZ).

48. Nelson MC, et al. (2012) Long-term vulnerability and resilience: Three examples from ar-chaeological study in the southwestern US and northern Mexico. Surviving Sudden Environ-mental Change, eds Cooper J, Sheets P (Univ Press of Colorado, Boulder, CO), pp 193–217.

49. IPCC (2012) Summary for policymakers. Managing the Risks of Extreme Events andDisasters to Advance Climate Change Adaptation: A Special Report of WorkingGroups I and II of the Intergovernmental Panel on Climate Change, eds Field CB,et al. (Cambridge Univ Press, Cambridge, UK), pp 3–22.

50. Me-Bar Y, Valdez F, Jr (2005) On the vulnerability of the ancient Maya society tonatural threats. J Archaeol Sci 32(6):813–825.

51. Torvinen A, et al. Transformation without collapse: Two cases from the AmericanSouthwest. Beyond Collapse: Archaeological Perspectives on Resilience, Revitalization,and Transformation in Complex Societies, ed Faulseit RK (Southern Illinois Univ Press,Carbondale, IL).

52. Lynnerup N (1998) The Greenland Norse: A Biological-Anthropological Study. Man &Society 24, Meddelelser om Grønland (Comm Sci Res Greenland, Copenhagen).

53. Minnis PE (1985) Social Adaptation to Food Stress: A Prehistoric SouthwesternExample (Univ of Chicago Press, Chicago).

54. Schollmeyer K (2009) Resource stress and settlement pattern change in the eastern Mim-bres area, southwest New Mexico. PhD dissertation (Arizona State University, Tempe, AZ).

55. Cannon MD (2001) Large mammal resource depression and agricultural intensification:An empirical test in the Mimbres Valley, New Mexico. PhD dissertation (University ofWashington, Seattle).

56. Sandor J, Gersper PL, Hawley JW (1990) Prehistoric agricultural terrace and soils in theMimbres area, New Mexico. World Archaeol 22(1):70–86.

57. HegmonM, Nelson MC (2007) In sync, but barely in touch: Relations between the Mimbresregion and the Hohokam regional system. Hinterlands and Regional Dynamics in theAncient Southwest, eds Sullivan AP, Bayman J (Univ of Arizona Press, Tucson, AZ), pp 70–96.

58. Van Gerven DP, Sheridan SG, eds (1994) The Pueblo Grande Project, Volume 6: TheBioethnography of a Classic Hohokam Population (Soil Syst Publ Archaeol, Phoenix).

59. Arneborg J, Lynnerup N, Heinemeier J (2012) Human diet and subsistence patterns inNorse Greenland AD c.980–AD c.1450: Archaeological interpretations. J NorthAtlantic 3(Special):119–133.

60. Enghoff IB (2003) Hunting, Fishing and Animal Husbandry at the Farm Beneath theSand, Western Greenland. Man & Society 28, Meddelelser om Grønland (Danish PolarCenter, Copenhagen).

61. McGovern TH (1985) Contributions to the paleoeconomy of Norse Greenland. ActaArchaeol 54:73–122.

62. Dugmore AJ, et al. (2012) Cultural adaptation, compounding vulnerabilities andconjunctures in Norse Greenland. Proc Natl Acad Sci USA 109(10):3658–3663.

63. Streeter R, Dugmore AJ, Vésteinsson O (2012) Plague and landscape resilience inpremodern Iceland. Proc Natl Acad Sci USA 109(10):3664–3669.

64. Abel N, Cumming DHM, Anderies JM (2006) Collapse and reorganization in the social-ecological systems: Questions, some ideas, and policy implications. Ecol Soc 11(1):17.

65. Madsen CK (2014) Pastoral settlement, farming, and hierarchy in Norse Vatnahverfi,South Greenland. PhD dissertation (University of Copenhagen, Copenhagen).






http://www.millenniumassessment.org


1

SUPPORTING INFORMATION

FILES 1-3

Climate Challenges, Vulnerabilities, and Food Security

Margaret C. Nelson, Scott E. Ingram, Andrew J. Dugmore, Richard Streeter, Matthew A. Peeples, Thomas

H. McGovern, Michelle Hegmon, Jette Arneborg, Keith W. Kintigh, Seth Brewington, Katherine A.

Spielmann, Ian A. Simpson, Colleen Strawhacker, Laura E.L. Comeau, Andrea Torvinen, Christian K.

Madsen, George Hambrecht, and Konrad Smiarowski

2


Supporting Information #1: Climate Challenges

Scott E. Ingram and Richard Streeter

Supporting information #1 describes the data and methods used to identify climate challenges in the U.S. Southwest and North Atlantic. The focal period in the Southwest is CE 1000 to 1500, 1000 to 1500 in Norse Greenland, and 900-1900 in Iceland. Different data and methods were used to identify the climate challenges in each region due to differences in climate regimes, data availability, and the subsistence strategies of the peoples considered.

Climate challenges, as discussed in the associated paper, are rare climate extremes. Climate extremes are defined as “The occurrence of a value of a weather or climate variable above (or below) a threshold value near the upper (or lower) ends of the range of observed values of the variable” (1). In both regions, we identify multi-year climate extremes because single-year events were likely accommodated by existing human adaptive and buffering strategies (2). Multi-year extremes are most likely to influence the productivity of the resources (plant and animal, wild and cultivated) humans rely on, exhaust existing buffering strategies, and prompt larger-scale and more archaeologically detectable behavioral responses.

U.S. Southwest

Paleoclimate Data

Precipitation and streamflow levels substantially influence food production (wild and cultivated) in the arid and semi-arid Southwest. Maize/corn (Zea mays L.) was the primary crop produced during the period of study and climate extremes, especially dry periods (droughts), decrease maize productivity (3-5) and can decrease food security. Cool temperatures that shorten the maize growing season can also decrease maize productivity at higher elevations in the Southwest. We do not consider cool temperature extremes in this study because the proxy data available lack sufficient spatial and temporal resolution. The relationship between the reconstructed temperature variable and maize productivity is also poorly understood. Identification of climate extremes (dry periods) is accomplished through the analysis of variation in tree-ring widths calibrated with modern, instrumental climate station data. The precipitation and streamflow reconstructions selected were closest to the primary settlement areas of each case. Climate varies substantially across the region (6) so tree-ring stations (sampling areas) closest to settlement areas are often the most representative of past climates in those areas. The Zuni (Cibola) and Salinas (Santa Fe) precipitation reconstructions were developed by Dean and Robinson (7). The Mimbres (Central Rio Grande) precipitation reconstruction was developed by Grissino-Mayer and colleagues (8). The Hohokam streamflow reconstruction (Lower Salt, Tonto, and Verde Rivers) was developed by Graybill (9). For a discussion of the strengths and weaknesses of tree-ring climate reconstructions see Fritts (10) and Dean (11). Methods of tree-ring based precipitation reconstruction are described by Rose and colleagues (12).

3

Identifying Climate Extremes We use a two-step method to identify multi-year climate extremes (dry periods). First, year-to-year variation in each precipitation reconstruction is smoothed with a centered nine-year-interval moving average. Climate in the Southwest, and most arid environments, is highly variable from year-to-year so multi-year periods of unvarying extreme low or high precipitation and streamflow are rare. Smoothing is necessary to identify the extreme periods within this variability. A nine-year interval is used as a compromise between shorter durations, which would not as faithfully represent trends in climate proxy data, and longer durations, which would obscure climate variation that would have been potentially meaningful for human behavior. A nine-year interval duration is supported by studies that have documented persistence (year to year similarity) in climate patterns on decadal scales in both the modern instrumental and proxy records (13-16). Moving averages have been used in a number of tree-ring based paleoclimatic studies to identify multi-year dry and wet periods (17-22). These interval averages accommodate but do not ignore anomalous years within an extreme period that likely do not end the extreme period. For example, a single wet year during a dry period would not end the dry period or necessarily replenish stored food reserves or soil moisture. Second, a threshold value is used to identify extreme periods within the reconstructions. We define extreme periods as those nine-year intervals in the first quartile (twenty-fifth percentile) of the distribution of interval averages of a precipitation reconstruction. Similar approaches have been used with standard deviation units by Dean (11) and percentile approaches to identify thresholds are currently used by the U.S. Drought Monitor (www.cpc.noaa.gov) and others to track drought severity across the U.S. (23-25). A first quartile threshold is arbitrary but assumed to represent periods with sufficient rarity to have substantially influenced resource productivity or human perceptions of productivity relative to typical conditions. We use the entire duration of the reconstructions (~1,500 years) to calculate the percentiles. The percentile value is assigned to year five of each moving nine-year interval. When four or fewer years of separation exist between the middle years of two identified extreme intervals, the intervals are merged into a single extreme period because the nine-year intervals substantially overlap. We ignore all extreme periods of four or fewer years. Shorter deleterious periods were likely effectively addressed by existing buffering strategies (e.g., storage, diet modification, trade). Identifying Rare Climate Challenges Rare climate challenges are identified among the frequently occurring extreme dry periods by calculating the number of years since a dry period of equal or greater duration had occurred. Table S1-1 contains the start and end dates, duration, and rarity of dry periods identified as rare climate challenges. Figure S1-1 identifies these periods in the context of all dry periods during the CE 1000 to 1500 focal period. The dry periods are considered rare because a dry period of equal or greater duration had not been experienced for at least 400 years. Existing human adaptive social and technological strategies for coping with dry period declines in food production would likely have been significantly challenged by these long and rare dry periods.

4

Table S1-1. Major climate challenges identified in the U.S. Southwest CE 1000 to 1500.

U.S. Southwest Cases

Climate challenges,

Years CE

Duration, years

Rarity: Years Since a Dry Period of Equal or Greater Duration1

Start of Precipitation or Streamflow Reconstruction, CE

Zuni 1133 - 1154 22 > 697 436 Salinas 1335 - 1351 17 > 456 879

Mimbres 1 1127 - 1147 21 > 505 622 Mimbres 2 1273 - 1295 23 > 651 622

Hohokam2 1 1338 - 1352 15 580 572 Hohokam2 2 1436 - 1452 17 678 572

Notes: 1 Years with the”>” sign indicate that the duration of the dry period exceeds the duration of any previously identified dry period in the tree-ring precipitation reconstruction. 2 Hohokam farmers relied on irrigation agriculture from a perennial river. A streamflow reconstruction is used to identify dry periods for this case.

5

Figure S1-1. Extreme Dry Periods in the U.S. Southwest from CE 1000 to 1500.

6

Notes: 1. Shaded areas show the dry periods and the rare climatic challenges discussed in the associated paper are identified with

arrows. 2. The Zuni (Cibola) and Salinas (Santa Fe) precipitation reconstructions were developed by Dean and Robinson (7). The

Mimbres (Central Rio Grande) precipitation reconstruction was developed by Grissino-Mayer and colleagues (8). The Hohokam streamflow reconstruction (Lower Salt, Tonto, and Verde Rivers) was developed by Graybill (9). All reconstructions were developed in association with the Laboratory of Tree-ring Research at the University of Arizona. Dry periods (shaded areas) were identified by Ingram using the methods described in this SI.

North Atlantic

The North Atlantic differs from the U.S. Southwest in climatic variability, paleoclimatic proxies, and the challenges faced by the predominantly pastoral farming and hunter gathering societies who lived there. Therefore a different methodology for identifying climate challenges was adopted. Our aim was to identify only the most significant climate challenges, where we could have reasonable certainty that they would have presented difficulties for the societies living there.

Paleoclimate data

The proxy records and reconstructions used are described in Table S1-2. Records were selected on the basis of geographical proximity to settlement areas, duration over the period of interest, and reconstructions which are relevant to the challenges outlined below. The main climatic challenges for societies living in the North Atlantic are 1) cold summer temperatures which severely inhibit fodder production crucial to animal provision (26), 2) increased sea ice occurrence, which decreases temperatures in northern Iceland and 3) increased storminess, which makes sea travel and fishing riskier, and may also decrease fodder production. Relative to the U.S. Southwest records, the North Atlantic climate proxies are of poor spatial and chronological resolution. Although documentary sources are an excellent source of climate information in this region (e.g., 27) we did not include them here as they are incomplete prior to CE 1500 and none exist for Greenland.

Identifying climate challenges in the North Atlantic

We identify climate extremes within the proxy records by selecting significant negative (cold) deviations from long-term means. These may be episodes of extended or exceptional (when considered on a multi-century scale) cold temperatures or increased storminess. We do this by identifying extended (> 3 years) deviations greater than one standard deviation from the mean. These can be considered climatic extremes.

In order to identify climatic challenges within the climate extremes we use additional criteria and sources of climate information. We considered the duration since an event of equal or greater magnitude (Table S1-2). We are most interested in identifying the first experienced occurrence of climatic extremes, as we assume these are most likely to have had the largest impact. Subsequent extremes, unless they involve significantly larger deviations from the new colder or stormier average conditions may have been less stressful, as societies are likely to have possessed enhanced traditional environmental knowledge in response to previous stresses. Given the generally poor spatial and chronological resolution of the records (chronological uncertainty for records was 5-20 years) we decided that climatic extremes must be visible in a minimum of two records to be classified as a climate challenge. This criterion decreases the likelihood of including changes that may have been only local in extent.

7

In addition to the proxy records identified in Table S1-2, we used information from climate models and additional climate proxies. It is known that not all proxy records (especially tree-ring based reconstructions) are sensitive to large magnitude climatic forgings. The VEI 7 magnitude eruption of Samalas, Indonesia in CE 1257 (28) results in the largest sulfate spike in the ice core record in the past 7000 years (37), but the proxy evidence here shows little cooling at this time (e.g., Figure S1-2 c, d). However, there are methodological reasons for this event to be under represented in tree ring records (32). When this volcanic forcing is incorporated into global circulation models there is a significant 2-3 year cooling in the northern hemisphere (31,32), and this cooling is recorded in documentary sources in Europe (38). On this basis we include this event in our challenges, despite limited evidence for it in the proxy records shown in Figure S1-2. Secondly, although Figure S1-2 shows some evidence of cooling in the early 1300s, noticeably in the increased incidence of sea ice north of Iceland, there is additional evidence that this time represents the start of a climatic shift in the North Atlantic. In Iceland there are changes in lake sediments (33) and in the Greenland region there is glaciological evidence of re-advances starting during this time (35).

Table S1-2 – Major climate challenges identified in the North Atlantic region AD 900-1900.

North Atlantic Cases

Climate challenges, Year CE1

Rarity – Years since an extreme of equal or greater duration2

Type of climate challenge

Justification

Iceland 1 1257-1260 >338 (29)3 Cool temperature extreme

Volcanically forced cooling due to eruption of Samalas, Indonesia in CE 1257 (28). Some indication in proxy records although uncertain due to low-resolution chronology (Figure 2, 29). Modeled scenarios show a large cooling (30, 31), which lasts for 2-3 years.

Greenland 1 1257-1260 NA Cool temperature

extreme See above.

Iceland 2 ca. 1310 90 (29)3; >589

(32)3 Cool temperature extreme and abundant sea ice

Start of change to Little Ice Age conditions and cooler temperatures in temperature reconstructions (29, 32). Shift to mean cooler temperatures recorded from CE 1300 in proxies in Iceland (33). Sea Ice proxies (34) see return of regular drift ice off the north coast of Iceland which had been largely absent for at least 700 years.

8

Greenland 2 ca. 1310 NA Cool temperature

extreme and abundant sea ice

See above, and (35).

Greenland 3 1421 NA Storminess extreme

and cool temperature extreme

Indicators of storminess in the North Atlantic show shift to increasing storminess around CE 1420, and a peak from 1451-1480 (Figure 2; 35,36). A large volcanic forcing around CE 1450 (30) suggests a cold period lasting 2-3 years.

Iceland 3 1640 102 (29)3; >1200

(34)3; 22 (32)3 Cool temperature extreme and abundant sea ice extreme

Significantly below average temperatures (29, 32) and highest levels of sea ice since start of record (34).

Notes: 1 Where challenges have a clearly defined duration this is shown, otherwise just the start date is identified. 2 Where challenges are based upon records which were not statistically tested for extremes such as the model scenarios and changes in lake sediments, NA is shown here. 3 Proxy record from which the rarity of the event is calculated.

9

Figure S1-2. Proxy records used to identify climatic challenges in the North Atlantic Region

Notes: Shaded areas show identified climatic challenges. (a) An alkonene based sea surface temperature reconstruction from core

MD99-2275 taken from off the north coast of Iceland (29). (b) A sea ice reconstruction using the IP25 biomarker from the same

core (MD99-2275). (c) A multi-proxy temperature reconstruction for the grid cell occupied by west Iceland, and (d) shows the

same reconstruction for the Eastern Settlement in Greenland (32). (e) Na+ anomalies from the GISP ice core, which are here

used as a proxy for storminess in the North Atlantic (35). Graphs a–c are relevant for understanding changes in Iceland, and

graphs d and e for understanding changes in the Eastern Settlement in Greenland.

10

References

1. IPCC (2012) Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation (Cambridge University Press, UK and New York).

2. Halstead P & O'Shea J (1989) Introduction: Cultural Responses to Risk and Uncertainty. Bad Year Economics: Cultural Responses to Risk and Uncertainty, eds Halstead P & O'Shea J (Cambridge University Press, Cambridge), pp 1-7.

3. Benson LV (2011) Factors Controlling Pre-Columbian and Early Historic Maize Productivity in the American Southwest, Part 1: the Southern Colorado Plateau and Rio Grande Region. Journal of Archaeological Method and Theory 18:1-60.

4. Muenchrath DA & Salvador RJ (1995) Maize Productivity and Agroecology: Effects of Environment and Agricultural Practices on the Biology of Maize. Soil, Water, Biology, and Belief in Prehistoric and Traditional Southwestern Agriculture, ed Toll HS (New Mexico Archaeological Council Special Publication. Volume 2. New Mexico Archaeological Council, Albuquerque), pp 303-333.

5. Shaw RH (1988) Climate Requirement. Corn and Corn Improvement, eds Sprague GF, Dudley JW (Agronomy Monograph 18, American Society of Agronomy, Madison, WI), Third Ed, pp 609-638.

6. Sheppard PR, Comrie AC, Packin GD, Angersbach K, Hughes MK (2002) The Climate of the US Southwest. Climate Research 21:219-238.

7. Dean JS, Robinson WJ (1978) Expanded Tree-Ring Chronologies for the Southwestern United States. (Laboratory of Tree-Ring Research, University of Arizona, Tucson).

8. Grissino-Mayer HD, Baisan CH, Swetnam TW (1997) A 1,373 Year Reconstruction of Annual Precipitation for the Southern Rio Grande Basin. Final report submitted to the Directorate of Environment, Natural Resources Division, Fort Bliss, Texas, for the Legacy Program. Available at http://web.utk.edu/~grissino/downloads/Legacy%20Final%20Report.pdf. Accessed November 5, 2010.

9. Graybill DA (1989) The Reconstruction of Prehistoric Salt River Streamflow in The 1982-1984 Excavations at Las Colinas: Environment and Subsistence, ed Graybill DA et al. (Cultural Resource Management Division, Arizona State Museum, University of Arizona, Tucson), pp 25-38.

10. Fritts HC (1976) Tree Rings and Climate (Academic Press, London, New York). 11. Dean JS (1988) Dendrochronology and Paleoenvironmental Reconstructions on the Colorado

Plateau. The Anasazi in a Changing Environment, ed Gumerman GJ (Cambridge University Press, Cambridge), pp 119-167.

12. Rose MR, Dean JS, & Robinson WJ (1981) The Past Climate of Arroyo Hondo New Mexico Reconstructed From Tree Rings (School of American Research Press, Santa Fe, NM).

13. Cayan DR, Dettinger MD, Diaz HF, Graham NE (1998) Decadal Variability of Precipitation over Western North America. J Clim 11:3148-3166.

14. Dettinger MD, Cayan DR, Diaz HF, Meko DM (1998) North–South Precipitation Patterns in Western North America on Interannual-to-Decadal Timescales. J Clim 11(12):3095-3111.

15. Gray ST, Jackson ST, and Betancourt JL (2004) Tree-Ring Based Reconstructions of Interannual to Decadal Scale Precipitation Variability For Northeastern Utah Since 1226 A.D. J Am Water Resour Assoc 40:947-960.

16. Grissino-Mayer HD (1995) The Climate and Fire History of El Malpais National Monument, New Mexico. Ph.D. dissertation (The University of Arizona, Tucson. University Microfilms, Ann Arbor).

17. Benson L, Petersen K, Stein J (2007) Anasazi (Pre-Columbian Native-American) Migrations During The Middle-12th and Late-13th Centuries--Were they Drought Induced? Clim Change 83:187-213.

18. Cordell LS, Van West CR, Dean JS, and Muenchrath DA (2007) Mesa Verde Settlement History and Relocation: Climate Change, Social Networks, and Ancestral Pueblo Migration. Kiva 72(4):391-417.

11

19. Ni F, Cavazos T, Hughes MK, Comrie AC, Funkhouser C (2002) Cool-Season Precipitation in the Southwestern USA Since AD 1000: Comparison of Linear and Nonlinear Techniques for Reconstruction. International Journal of Climatology 22:1645-1662.

20. Parks JA, Dean JS, Betancourt JL (2006) Tree Rings, Drought, and the Pueblo Abandonment of South-Central New Mexico in the 1670s. Environmental Change and Human Adaptation in the Ancient American Southwest, ed Dean JS (The University of Utah Press, Salt Lake City), pp 214-227.

21. Van West CR, Grissino-Mayer HD (2006) Dendroclimatic Reconstruction. Fence Lake Project: Archaeological Data Recovery in the New Mexico Transportation Corridor and First Five-Year Permit Area, Fence Lake Coal Mine Project, Catron County, New Mexico, ed Huber EK, Van West CR (SRI Press, Tucson), pp 33.31-33.

22. Woodhouse CA (2001) A Tree-Ring Reconstruction of Streamflow for the Colorado Front Range. J Am Water Resours Assoc 37(3):561-569.

23. Hirschboeck KK, Meko DM (2005) A Tree-Ring Based Assessment of Synchronous Extreme Streamflow Episodes in the Upper Colorado and Salt-Verde-Tonto River Basins: A Collaborative Project between The University of Arizona's Laboratory of Tree-Ring Research and The Salt River Project. Available at http://www.ltrr.arizona.edu/kkh/SRP/SRP.LTRR.Press.info.05.slides.pdf. Accessed November 5, 2010.

24. Steinemann A, Hayes MJ, Cavalcanti LFN (2005) Drought Indicators and Triggers. Drought and Water Crises: Science, Technology, and Management Issues, ed Wilhite D (CRC Press, Boca Raton, FL), pp 71-92.

25. Smakhtin VU (2001) Low Flow Hydrology: A Review. Journal of Hydrology 240:147-186. 26. Bergthorsson, P (1985) Sensitivity of Icelandic Agriculture to Climatic Variations. Climatic Change

7:111-127 27. Ogilvie AEJ (1984) The past climate and sea-ice record from Iceland, Part 1: Data to A.D. 1780. Clim

Change 6:131-152 28. Lavigne F, Degeai J-P, Komorowski J-C, et al. (2013) Source of the great A.D. 1257 mystery eruption

unveiled, Samalas volcano, Rinjani Volcanic Complex, Indonesia. Proc Natl Acad Sci U S A 110: 16742-16747.

29. Sicre MA, Hall IR, Mignot J, et al. (2011) Sea surface temperature variability in the subpolar Atlantic over the last two millennia. Paleoceanography 26.

30. Mann ME, Fuentes JD and Rutherford S. (2012) Underestimation of volcanic cooling in tree-ring-based reconstructions of hemispheric temperatures. Nat Geosci 5: 202-205.

31. Schurer AP, Hegerl GC, Mann ME, et al. (2013) Separating Forced from Chaotic Climate Variability over the Past Millennium. J Clim 26: 6954-6973.

32. Mann ME, Zhang Z, Rutherford S, et al. (2009) Global Signatures and Dynamical Origins of the Little Ice Age and Medieval Climate Anomaly. Science 326: 1256-1260.

33. Geirsdóttir Á, Miller GH, Larsen DJ, et al. (2013) Abrupt Holocene climate transitions in the northern North Atlantic region recorded by synchronized lacustrine records in Iceland. Quat Sci Rev 70: 48-62.

34. Masse G, Rowland SJ, Sicre M-A, et al. (2008) Abrupt climate changes for Iceland during the last millennium: Evidence from high resolution sea ice reconstructions. Earth Planet Sci Lett 269: 564-568.

35. Dawson AG, Elliott L, Mayewski P, et al. (2003) Late-Holocene North Atlantic climate 'seesaws', storminess changes and Greenland ice sheet (GISP2) palaeoclimates. Holocene 13: 381-392.

36. Dugmore AJ, Borthwick DM, Church MJ, et al. (2007) The role of climate in settlement and landscape change in the north Atlantic islands: An assessment of cumulative deviations in high-resolution proxy climate records. Hum Ecol Interdiscip J 35: 169-178.

37. Gao C, Robock A and Ammann C. (2012) Volcanic forcing of climate over the past 1500 years: An

12

improved ice core-based index for climate models (vol 113, D2311, 2008). J Geophys Res Atmos 117.

38. Stothers RB (2000) Climatic and demographic consequences of the massive volcanic eruption of 1258. Clim Change 45:361-374.

13


Supporting Information #2: Cases and Vulnerability Load

Margaret C. Nelson, et al.1

Supporting Information #2 includes the archaeological and historical evidence used to support codes

assigned to each of the eight variables contributing to vulnerability to food security in each case. A case

is delineated by the time frame around a climate challenge (see SI#1) in each of the regions included in

this study (Fig. S2-1 and Fig. S2-2). We focus on four regions in the US Southwest and three regions in

the North Atlantic. The North Atlantic and US Southwest cases are used because in both macro-regions,

teams of researchers have a decade-long history of studying the same kinds of questions about human-

environment interaction over long time frames (500 to over 1000 years). These studies provide the

basis for the evidence documented in SI#1, SI#3 and below. The collaboration of researchers in both

macro-regions provided an opportunity to identify patterns of human-environment-climate interactions

that supersede the specifics of cultural tradition and of climate and environmental conditions. Coding

was done by all co-authors in face-to-face meetings.

Figure S2-1: Four regions in the US Southwest used for this study: Hohokam, Zuni, Salinas, Mimbres

1 All authors to the article as listed on the cover page of this SI document

14

Figure S2-2: Three North Atlantic Islands used for this study: Greenland, Iceland, and Faroe Islands

Vulnerability can encompass a wide range of factors, but for the purpose of this study we identified

eight variables, three population-resource and five social, to quantify the vulnerability of people to food

shortage (1). These are variables that we know contribute to food security and that can be observed in

the archaeological and historical record. They allow calculation of “vulnerability load,” which can be

thought of as the weight of human-created factors in the potential for shortfalls in food.

The codes for vulnerability load are 1 = no contribution to vulnerability, 2 = limited contribution to

vulnerability, 3 = more than limited but not substantial, 4 = substantial contribution to vulnerability load.

Codes 1 and 4 define the ends of a continuum for the contribution of each variable to vulnerability load.

Between these ends, Code 2, “limited contribution to vulnerability,” is recognized by mixed evidence for

potential contribution to shortfall that is closer to “no vulnerability” than to substantial vulnerability.

Code 3, “more than limited but not substantial,” is recognized by multiple lines of evidence that are

closer to substantial contribution than to no contribution. For example, among the population-resource

variables, depletion of one minor food resource would be coded as 2, “limited contribution to

vulnerability,” while depletion of a key resource would be coded as 3. As a further example, in the social

realm, mobility was coded 2 if either inside region movement or mobility to areas outside the region

was restricted in any way – by lack of adequate transport, by laws, social norms, or conflict. Mobility

would be coded 3 if these conditions limited both inside and outside movement. With strong evidence

that people did not move, the variable would be coded 4.

15

Table S2-1. Codes and supporting archaeological and historical evidence are presented by case. Dates in

red are the start dates for the climate challenges; vulnerability loads are calculated for the period

immediately before the date in red by adding the eight numbers for each case.

Iceland 1 1257 Code Evidence

Pop/ Resource Domain

1. Plenty of food 1

Zooarchaeological data indicate adequate food supply (2). Limited erosion and environmental degradation in North Iceland, some similar patterns in south Iceland but also evidence of stable or declining levels of erosion (3-8). Barley production on-going (9).

2. Diversity of Available Food 1

Zooarchaeological data indicate wide range of animals, marine resources (2). Full range of domestic animals and farming practices recognizable (2).

3. Resource Depression 2

Geomorphic evidence of ongoing erosion in lowland areas in south Iceland that limit people’s options (3, 6, 8, 10). People could not gain any more resource productivity from manuring; no extra manuring was required for existing level of productivity (11). Woodland clearing had progressed across landscape (12). Grassland ecosystems slowly replacing woodland (13, 14).

Social Domain

4. Outside area network 1

Historical records document connections between communities and direct communication with Scandinavia, British Isles, and Greenland; the Norwegian kings tied together Scandinavia, the North Atlantic and the British Isles (15). Interactions among communities within Iceland in different zones (16).

5. Storage 1

Evidence of hay storage - excavated houses with large hay barrels (17). Evidence for dried fish (18, 19). Some resources could be stored for several years (19).

6. Mobility 2

Internal mobility was possible and Iceland is big enough to have diverse options; but leaving the island would have been challenging. Movement was more constrained by social structure (textual evidence – 16).

7. Equal access to food 2

Presence as hreppur - way to mediate food stress by creating obligations between classes; rules/laws determine who has access and in what order, which creates vulnerabilities for under class (20). Zooarchaeological data for unequal access to food (21). Presence of food resources inland indicating wide distribution of resources (19).

8. Barriers to access 2 Historic evidence for conflict limiting access to highland zone (22).



1. Plenty of food 2

Historic annals of some starving (23). Neonatal fatalities of sheep indicating loss of lambs (24, 25). Stress in provisioning indicated by increased seals in diet (26).


Zooarchaeological data indicate wide range of animals, marine resources (2). Full range of domestic animals and farming practices recognizable (2). Shift away from cows and goats and declines in barley local production but barley was being imported (18).


Geomorphic evidence of ongoing erosion in lowland areas in south Iceland that limited people’s options (3, 6, 8, 10). People could not gain any more resource productivity from manuring; no extra manuring was required for existing level of productivity (11). Woodland clearing had progressed across landscape (12). Grassland ecosystems slowly replacing woodland (13, 14).

Social Domain

16


Historical records document connections between communities and direct communication with Scandinavia, British Isles, and Greenland. The Norwegian kings tied together Scandinavia, the North Atlantic and the British Isles (15). Material evidence for interactions among communities within Iceland in different resource zones (16).

5. Storage 1

Evidence of hay storage - excavated houses with large hay barrels (17). Evidence for dried fish (18, 19). Some resources could be stored for several years (19).

6. Mobility 2

Internal mobility was possible and Iceland is big enough to have diverse options; but leaving island would have been challenging; movement more constrained by social structure (textual evidence – 16).


Presence as hreppur - way to mediate food stress by creating obligations between classes; rules/laws determine order of access, which creates vulnerabilities for under class (27). Zooarchaeological data for unequal access to food (21). Presence of food resources inland indicating wide distribution of resources (19).

8. Barriers to access 2

Conflict of the previous century was over (16) but sea ice had developed in summer, which limited access to many marine resources (28-30).



1. Plenty of food 2

Increasing levels of soil erosion (8, 10, 31). Historic records document people complaining about hunger, food system not breaking down, but some stress (23).


Increased sea ice (30, 32, 33); high levels of sea ice off North Coast (34). Some resources not available, but other new resources available – increased food imports offset cessation of cereal cultivation (16).


Soil erosion high but not at highest level recorded (8, 35), which limited people’s options. Fodder and pasture productivity maintained but could not be increased by manuring (11). River sedimentation caused loss of access to certain fish (13, 36). Some glacial buildup started that reached maximum in late 17th century or early 18th century (37-39); farms threatened by encroaching ice (40, 41). Zooarchaeological data indicate decline in proportion of all domestic stock in relation to fish, so less meat and dairy in diet (42).

Social Domain


Some increase in trade/connections outside the island (16), but controlled by monopoly, so only some had security from trade; historical documents of monopoly (43, 44). Imported objects present but not common in archaeological sites; presence correlates with wealth (45).

5. Storage 2

Historic documents for erosion of capacity for storage based on increasing numbers of low status people, who moved often; with tenants moving often, infrastructure for storage was not a priority for them. Wealth was a major conditioner of ability and interest in storing (16). Home fields still productive (46) but hay not storable in field anymore (20).

6. Mobility 2

Evidence for commitment to specific place and residential stability by only small group of elites; tenants were forced to be mobile (47); lots of abandoned farms. Difficulty of leaving island because of limited access to boats (16).


The same condition of social inequality as earlier, but under conditions of less available food suggests worse conditions for some. In the cannons, some people were referred to as the expendables; their access to food was limited by their class (43).


Increased erosion and growth in ice limited transportation routes (23, 29). Glacial expansion caused lakes to be dammed and caused flooding in a few places (48).

Greenland 1 1257 Code Evidence


1. Plenty of food 1 Paleoenvironmental, zooarchaeological, and bio-physical assessments indicate an adequate resource base (49-53).

17


Farming decreased, seal hunting increased based on bioarchaeological isotope data (54, 55) and zooarchaeological data showing increase in seal bones (49, 53).


Increases in agro-pastoral activities over time contributed to accelerated soil erosion (55). Although soil-erosion indicators decreased after 1230, erosion remained at a high level (55). Creation of hay fields may have caused problems in terms of nutrients available in soils (when compared to landnam); additional effort in soil managements may have been at limits of investments based on paleoenvironmental data (56).

Social Domain


Not much movement of people and no food resources available from outside, although Greenlanders maintained contact with Iceland and Norway (57).

5. Storage 1 Architectural indications (elevated stone built store houses, large wooden barrels in storerooms) of adequate storage capability (58, 59).

6. Mobility 3

Not much movement of people, although isotope data show some movement between Greenland settlements (57) and some to Norway. Historical evidence of Greenlanders present in Bergen (GHM III) and archaeological evidence of contact – imports from Norway.


Isotopic data show differentiation in access to food by wealth/class (54). Architectural evidence (sizes of barns, byres, etc.) indicates different capacity to acquire food (58, 59).


Long, dangerous sails from mid- and inner fjord settlement to seal hunting grounds on the outer coast based on archaeological and zooarchaeological data (4).

Greenland 2 1310 Code Evidence


1. Plenty of food 1 Paleoenvironmental, zooarchaeological, and bio-physical assessments indicate adequate resource base (49-53).


Further increases in emphasis on seal hunting extended the pattern of decreased farming. Increased seal hunting based on bioarchaeological isotope data (54) and zooarchaeological data showing increase in seal bones (49, 53).


Increases in agro-pastoral activities over time contributed to accelerated soil erosion (55). Although soil-erosion indicators decreased after 1230, erosion remained at a high level (55). The creation of hay fields may have been causing problems in terms of nutrients available in soils (when compared to landnam); additional effort in soil management may have been at limits of investments based on paleoenvironmental data (56).

Social Domain


Submitting to the Norwegian crown to secure contacts could be a sign of declining contacts in a period before 1261 (57). Ecclesiastical contacts Gardar – Bergen/Nidaros in Norway (60).

5. Storage 1 Architectural indications (elevated stone built store houses, large wooden barrels in storerooms) of adequate storage capability (58, 59).

6. Mobility 4 Mobility further declined: aftermath of 1257 volcanic eruption impacted local intra-island travel and connections with Norway may have declined (see above).


Isotopic data show differentiation in access to food by wealth/class (54). Architectural evidence (sizes of barns, byres, etc.) indicates different capacity to acquire food (58, 59).


Long, dangerous sails from mid- and inner fjord settlement to seal hunting grounds on the outer coast based on archaeological and zooarchaeological data (4).

Greenland 3 1421, 1450 Code Evidence

Population/Resource Domain

1. Plenty of food 1 Even though high dependence on seals, there was enough for the people who were there. No skeletal evidence of hunger (52).


Diversity of food resources decreasing – many more seals – much less animal husbandry based on isotope data (54) and zooarchaeology (49, 53). Substantial decline of agro-pastoral activities (50, 51, 55).

18


Harbor seal completely gone. Lots of soil had been degraded (55). Irrigation systems and shieling intensification as an effort to continue hay production which was declining in the period 1260 – 1350 (51, 61). Increase in summer sea ice contributed to reduced pasture productivity.

Social Domain


Last known royal-controlled navigation between Norway and Greenland terminated around 1420 but some Icelanders in Greenland around 1400 (via Norway) (62). Last bishop residing in Greenland died in 1378; lack of replacement indicates decline in contact with Norway (60).

5. Storage 2 Architectural indications (elevated stone built store houses, large wooden barrels in storerooms) of storage (58, 59) though fewer kinds of resources to store.

6. Mobility 4

Less internal and external movement -- Western Settlement depopulated before 1400 (57), hunting trips to the North may have ended due to less sale of Arctic commodities in N Europe and the ending of navigation between Greenland and Norway (57).


Elite farms still have cattle in their byres – smaller farmers dependent on the marine resources (seal) – still fewer differences in diet than in earlier period – everyone depended on seal (54, 58).


Long, dangerous sails from mid- and inner fjord settlement to seal hunting grounds on the outer coast based on archaeological and zooarchaeological data (4) and settlement pattern data (58). As seal had become so essential, this barrier was greater than previously.

Faroes 1257 Code Evidence


1. Plenty of food 2

Though actual evidence is limited, there should have been enough food, given the (probably) low population level (63). However, because of the socio-economic changes occurring prior to the shock (64, 65), there might have been significantly increased pressures on the terrestrial resources.


Includes domestic livestock, marine and terrestrial wild resources, and at least some barley production (13, 14, 66-68).


Evidence is limited, but there is no indication of significant resource depression at this point (68, 69), though there was pressure on terrestrial resources.

Social Domain


Some trade and other connections to the outside world (NE Europe) (70, 71). There was certainly contact and exchange between communities, though no substantial differences in resources or environmental conditions between island communities in the Faroes.

5. Storage 2

Some storage in the form of grain (barley), livestock, and preserved foodstuffs (eggs, air-dried fish & meat). Evidence for preservation comes mostly from ethnographic record from later centuries (e.g. 72, 73) and known practices elsewhere in the Norse regions.

6. Mobility 3

Investment in place; severe restrictions on relocation of settlements because of the mountainous topography (63). Heavy reliance on boats for transportation and movement between islands and abroad.


As in other Norse societies (74), access was limited by social class. Mahler argued that access to resources was tied to land ownership in particular (64).


Some slight differences in resource availability depending on settlement location, but these were generally dealt with through exchange. Boats were a requirement for accessing deep-sea fish (75). There were physical hazards associated with fowling (73).

Mimbres 1 1127 Code Evidence


1. Plenty of food 2 Largest population (76-79). Riparian depletion in Mimbres Valley but not in eastern Mimbres (78, 80).


Large game limited -- Artiodactyl depletion well before this time (80). Some riparian depletion (78).

19


Artiodactyl populations reduced (81). Riparian habitat reduced in Mimbres Valley (78). Some soils impacted in Mimbres Valley (82).

Social Domain

4. Outside area network 4 Imported ceramics are extremely rare (83), though some Mesoamerican imagery (84, 85)

5. Storage 2

Ethnographic research on Puebloan agriculture records that people tried to keep 2-3 years of stored maize, the primary stored food resource. (86-88). Some households lacked formal storage space (89, 90).

6. Mobility 2 Place-focused settlement along major rivers, long site occupation histories (e.g., 91).


Possible better access for first comers, based on archaeological data (92, 93) and reference to ethnography (94, 95).

8. Barriers to access 1 No physical barriers evident.

Mimbres 2 1273 Code Evidence


1. Plenty of food 1 Population low relative to available resources and earlier periods (77-79).


Large game limited -- Artiodactyl depletion well before this time (80). Riparian recovery (78).

3. Resource Depression 1 Artiodactyl populations may have rebounded somewhat (96-97).

Social Domain

4. Outside area network 1 Variety of non-local ceramics and architecture (98, 99).

5. Storage 2

Ethnographic research on Puebloan agriculture records that people tried to keep 2-3 years of stored maize, the primary stored food resource. (86-88). Formal storage areas are rare (98).

6. Mobility 1 Residential mobility relatively high as evident in house architecture (98, 100). Many pan-regional connections evident in ceramic diversity (98, 99).

7. Equal access to food 1 All house units similar in size and composition of household features (98).

8. Barriers to access 1 No physical barriers.

Zuni 1133 Code Evidence


1. Plenty of food 1

No evidence that people faced food shortages in the region and population estimates (101, 102) for well-documented areas suggest that population levels likely would have been sustained easily, given the distribution of agricultural land.


In general, evidence from excavated sites suggests that people used a diversity of agricultural foods, wild plants, and animals, though the availability of large animals was somewhat reduced by this period (103, 104).


Little zooarchaeological or paleoethnobotanical evidence of resource depression in the region at this time, though ratios of artiodactyls to other animals in excavated assemblages suggest that they were somewhat reduced from earlier periods (103, 104).

Social Domain


Ceramic evidence suggests that connections to areas outside of the broader Zuni/Cibola region were limited (numbers of non-local sherds are miniscule) but there is some ceramic (103) and obsidian (105) evidence of connections that span the Colorado Plateau south to the Mogollon Highlands, an area marked by different environmental conditions.

5. Storage 2

Extensive evidence of storage in the form of dedicated rooms, pits in habitation rooms, and jars with seeds and other contents suggesting that people in the area were regularly able to store food (103, 104). Ethnographic research on Puebloan agriculture records that people tried to

20

keep 2-3 years of stored maize, the primary stored food resource (86-88).

6. Mobility 1

The Zuni region was marked by short-lived sites and a high degree of residential mobility as suggested by settlement data, the frequencies of dated ceramic types, and tree ring dates from several regions (101, 102, 106).


No evidence that any segment of society controlled access to food resources in the study area, in particular due to the high frequency of mobility (101, 102).


No evidence that substantial barriers to access to food resources existed. Settlement was extensive during this period (101, 102) and people residing across the region used a common suite of resources.

Salinas 1335 Code Evidence


1. Plenty of food 1 No evidence for imbalance.


In general, paleoethnobotanical and zooarchaeological evidence from excavated sites suggests that people used a diversity of agricultural foods, wild plants, and animals, though the availability of large animals was somewhat reduced by this period (107). Pinyon was an important complement to agricultural and hunted foods (108).

3. Resource Depression 1 Not until 15th century (107).

Social Domain


Little evidence for outside area connections, with only a small amount Rio Grande glaze wares (109-111).

5. Storage 2 Ethnographic research on Puebloan agriculture records that people tried to keep 2-3 years of stored maize, the primary stored food resource (86-88).

6. Mobility 4 As agriculturalists people were tethered to the water table and did not move away from it (79).

7. Equal access to food 1 No evidence for differential distribution or access to resources.


Localized areas of high water table limited where people could acquire and grow key resources (79).

Hohokam 1 1338 Code Evidence


1. Plenty of food 2 Population was large as documented with settlement data (112).


Zooarchaeological data show decline in terrestrial game, which led to increased dependence on riparian resources (James 2003). Artiodactyls limited (113).

3. Resource Depression 2 Zooarchaeological data indicate impact on game resources (114-116).

Social Domain


Balkanized system – widespread ballcourt system that had supported market exchange over large region was gone (117). Population was concentrated in river valleys (112, 117).

5. Storage 2

Ethnographic research on Puebloan agriculture records that people tried to keep 2-3 years of stored maize, the primary stored food resource (86-88).

6. Mobility 4 Canal infrastructure tied people to canals, restricting mobility (118) relative to other regions of the Southwest (106).


Control of irrigation system by some; platform mounds for ceremony and possibly elite residence were at headgates to canals controlling water and land (119-121).

8. Barriers to access 3 Canals delimited primary agricultural areas (122-124).

Hohokam 2 1436 Code Evidence

Pop/ Resource

21

Domain

1. Plenty of food 3 Population was large as document with settlement data (112). In addition, the primary source of food, the fields fed by the canal system, were decreasing in extent (123, 124).


Zooarchaeological data show decline in terrestrial game, which led to increased dependence on riparian resources (115). Artiodactyls limited (113).


Zooarchaeological data indicate impact on game resources (114-116). In addition, soil degradation has been documented along the Salt River (125).

Social Domain


Balkanized system – widespread ballcourt system that supported market exchange over large region was gone (117) and population was concentrated in river valleys (112, 117).

5. Storage 2 Ethnographic research on Puebloan agriculture records that people tried to keep 2-3 years of stored maize, the primary stored food resource (86-88).

6. Mobility 4 Canal infrastructure tied people to canals, restricting mobility (118) relative to other regions of the Southwest (106).


Control of irrigation system by some; platform mounds for ceremony and possibly elite residence were at headgates to canals controlling water and land (119-121).

8. Barriers to access 3 Canals delimited primary agricultural areas (122-124).

References 1. Turner BL, III, et al. (2003) A framework for vulnerability analysis in sustainability science. Proc Natl

Acad Sci USA 100(4):8074-8079. 2. Smiarowski K, et al. (in press) Zooarchaeology of the Scandinavian Settlements in Iceland and

Greenland: Diverging Pathways. Oxford Handbook of Zooarchaeology, eds Albarella U, Russ H, Vickers K, Viner-Daniels S (Oxford Univ. Press, Oxford).

3. Dugmore AJ, et al. (2006) An over-optimistic pioneer fringe? Environmental perspectives on medieval settlement abandonment in Thorsmork, south Iceland. The Dynamics of Northern Societies, eds Arneborg J, Grønnow B (National Museum, Copenhagen), pp 333-344.

4. Dugmore AJ, et al. (2007) Abandoned farms, volcanic impacts, and woodland management: Revisiting the ‘Pompeii of Iceland’.’ Arctic Anthropol 44(1):1-11.

5. Mairs K-A (2007) Islands and human impact: Under what circumstances do people put unsustainable demands on island environments? Evidence from the North Atlantic. PhD dissertation (Univ. of Edinburgh, Edinburgh).

6. Mairs K-A, Church MJ, Dugmore AJ, Sveinbjarnardóttir G (2006) Degrees of success: Evaluating the environmental impacts of successful long term settlement in south Iceland. The Dynamics of Northern Societies, eds Arneborg J, Grønnow B (Publications from the National Museum, Studies in Archaeology and History, Volume 10. Copenhagen), pp 363-372.

7. Ólafsdóttir R, Guómundsson HJ (2002) The Holocene land degradation and climatic change in northeastern Iceland. The Holocene 12(2):159-167.

8. Streeter RT, Dugmore AJ (2014) Late-Holocene land surface change in a coupled social-ecological system, southern Iceland: A cross-scale tephrochronology approach. Quat Sci Rev 86:99-114.

9. Trigg H, Bolender D, Johnson K, Patalano M, Steinberg J (2009) Note on barley found in dung in the lowest levels of the Farm Mound Midden at Reynistaður, Skagafjörður Iceland. Archaeologia Islandica 7:64-72

10. Dugmore AJ, Gisladóttir G, Simpson IA, Newton AJ (2009) Conceptual models of 1200 years of Icelandic soil erosion reconstructed using tephrochronology. Journal of the North Atlantic 2(1):1-18.

22

11. Adderley WP, Simpson IA, Vésteinsson O (2008) Local scale adaptations: a modeled assessment of soil, landscape, microclimatic and management factors in Norse home-field productivities. Geoarchaeology 23:500-527.

12. Church M, et al. (2007) Charcoal production during the Norse and Early Medieval periods in Eyjafjallahreppur, Southern Iceland. Radiocarbon 49(2):659-672.

13. Lawson IT, et al. (2006) Human impact on freshwater environments in Norse and early medieval Myvatnssveit, Iceland. Dynamics of Northern Societies. Proceedings of the SILA/NABO conference on Arctic and North Atlantic Archaeology, Copenhagen, May 10th-14th, 2004, eds Arneborg J, Grønnow B (Publications of the National Museum. Studies in Archaeology and History 10, Copenhagen), pp 375-382.

14. Lawson IT, et al. (2007) Environmental impacts of the Norse settlement: Palaeoenvironmental data from Mývatnssveit, northern Iceland. Boreas 36(1):1-19.

15. Gad F (1967) Grønlands Historie vol I Indtil 1700 (København, Copenhagen). 16. Karlsson G (2000) Iceland’s 1100 years: The History of a Marginal Society (C. Hurst, London). 17. Berson B (2002) A contribution to the study of the Medieval Icelandic farm: The Byres.

Archaeologia Islandica 2:34-60. 18. Harrison RB (2013) World systems and human ecodynamics in medieval Iceland: Gásir in

Eyjafjordur. PhD dissertation (City Univ. of New York, New York, NY). 19. McGovern TH, Perdikaris S, Einarsson Á, Sidell J (2006) Coastal connections, local fishing, and

sustainable egg harvesting: Patterns of Viking Age inland wild resource use in Mývatn District, Northern Iceland. Environmental Archaeology 11:187-205.

20. Eggertsson T (1998) Sources of risk, institutions for survival, and a game against nature in premodern Iceland. Explor Econ Hist 35:1–30.

21. Ólafsson G, McGovern TH, Smith KP (2006) Outlaws of Surtshellir Cave: The underground economy of Viking Age Iceland. Dynamics of Northern Societies, Proceedings of the SILA/NABO conference on Arctic & North Atlantic Archaeology 2004, eds Arneborg J, Grønnow B (National Museum of Denmark, Copenhagen), pp 395-407.

22. McGrew J (1970) Sturlunga Saga (Twayne Publishers, New York). 23. Ogilvie AEJ (1991) Climatic changes in Iceland A.D. c. 1500 to 1598. The Norse of the North Atlantic,

ed Bigelow GF (Acta Archaeologica 61(1900), Munksgaard, Copenhagen, Denmark), pp 233–251. 24. Amorosi T (1992) Climate impact and human response in NE Iceland: Archaeological investigations

at Svalbard 1986-88. Norse and Later Settlements and Subsistence in the North Atlantic, eds Morris CD, Rackham JD (Univ. of Glasgow, Dept. of Archaeology, Glasgow), pp 101-127.

25. Gísladóttir G, et al. (2013) The Svalbarð Project. Archaeologia Islandica 10:69-103. 26. McGovern TH, Harrison R, Smiarowski K (2014) Sorting sheep & goats in Medieval Iceland and

Greenland: Local subsistence or world system? Long-Term Human Ecodynamics in the North Atlantic: An Archaeological Study, eds Harrison R, Maher R (Lexington, Lanham, MD), pp 153-175.

27. Hastrup K (1990) Nature and Policy in Iceland 1400–1800: An Anthropological Analysis of History and Mentality (Clarendon, Oxford).

28. Andrews JT, et al. (2009) A robust, multisite Holocene history of drift ice off northern Iceland: Implications for North Atlantic climate. The Holocene 19(1):71–77.

29. Geirsdóttir Á, Miller GH, Larsen DJ, Ólafsdóttirs (2013) Abrupt Holocene climate transitions in the northern North Atlantic region recorded by synchronized lacustrine records in Iceland. Quat Sci Rev 70:48-62.

30. Miller G, et al. (2012) Abrupt onset of the Little Ice Age triggered by volcanism and sustained by sea-ice/ocean feedbacks. Geophys Res Lett 39(2):1-5.

23

31. Simpson IA, Dugmore AJ, Thomson A, Vésteinsson O (2001) Crossing the thresholds: human ecology and historical patterns of landscape degradation. Catena 42(2-4):175–192.

32. Ogilvie AEJ, Jónsson T (2001) “Little Ice Age” research: A perspective from Iceland. Clim Change 48:9–52.

33. Ogilvie AEJ, Jónsdóttir I (2000) Sea ice, climate, and Icelandic fisheries in historical times. Arctic 53(4):383–394.

34. Massé G, et al. (2008) Abrupt climate changes for Iceland during the last millennium: Evidence from high resolution sea ice reconstructions. Earth Planet Sci Lett 269:565–569.

35. Dugmore AJ, Buckland PC (1991) Tephrochronology and late Holocene soil erosion in South Iceland. Environmental Change in Iceland, eds Maizels J, Caseldine C (Kluwer, Dordrecht), pp 147-159.

36. Lawson IT, et al. (2005) Historical ecology on Sandoy, Faroe Islands: Palaeoenvironmental and archaeological perspectives. Hum Ecol 33:651-684.

37. Kirkbride MP, Dugmore AJ (2006) Responses of mountain ice caps in central Iceland to Holocene climate change. Quat Sci Rev 25(13/14):1692-1707.

38. Kirkbride MP, Dugmore AJ (2008) Tephrochronological dating of glacier advances AD 410-1947 in Southern Iceland. Quaternary Research 70(3):398-411.

39. Larsen DJ, Miller GH, Geirsdóttir Á, Thordarson T (2011) A 3000-year varved record of glacier activity and climate change from the proglacial lake Hvítárvatn, Iceland. Quat Sci Rev 30:2715-2731.

40. Grove JM (1987) Glacier fluctuations and hazards: Geogr J 153(3):351-367. 41. Thorarinsson S (1956) The thousand years struggle against ice and fire. Miscellaneous Papers, No.

14 (Museum of Natural History, Dept. of Geology and Geography, Reykjavik). 42. Perdikaris S, McGovern TH (2008) Codfish and kings, seals and subsistence: Norse marine resource

use in the North Atlantic. Human Impacts on Marine Environments, eds Torben R, Erlandson J (Historical Ecology Series, Univ. of California, Los Angeles Press, Los Angeles), pp 157-190.

43. Gunnarsson G (1983) Monopoly Trade and Economic Stagnation. Studies in the Foreign Trade of Iceland 1602–1787 (Ekonomisk–Historiska Fo¨reningen, Lund).

44. Gunnarsson G (1987) Upper bothith I´saland: Einokunarverslun og ı´slenskt samfe´lag 1602–1787 (Bókaútgáfan Örn og Örlygur, Reykjavı´k).

45. Sveinbjörnsdóttir, G (1996) Pottery found in Excavations in Iceland (National Museum of Iceland, Reykjavik).

46. Eggertsson T (1992) Analyzing institutional successes and failures: A millennium of common mountain pastures in Iceland. Int Rev Law Econ 12:423–437.

47. Vasey DE (1996) Population regulation, ecology, and political economy in preindustrial Iceland. Am Ethnol 23(2):366-392.

48. Dugmore A, Vesteinsson O (2012) Black sun, high flame and flood; volcanic hazards in Iceland. Surviving Sudden Environmental Change: Answers from Archaeology, eds Cooper J, Sheets P (Univ. Press of Colorado, Boulder), pp 67-91.

49. Enghoff IB (2003) Hunting, fishing and animal husbandry at The Farm Beneath the Sand, Western Greenland. Meddelelser om Grønland, Man & Society 28:104.

50. Ledger PM, Edwards KJ, Schofield JE (2013) Shieling activity in the Norse Eastern Settlement: Palaeoenvironment of the ‘Mountain Farm’, Vatnahverfi, Greenland. The Holocene 23(6):810-822.

51. Ledger PM, Edwards KJ, Schofield JE (2014) Vatnahverfi: A green and pleasant land? Palaeoecological reconstructions of environmental and land-use change. In the Footsteps of Vebæk—Vatnahverfi Studies 2005-2011. Journal of the North Atlantic. Special Volume 6:29–46.

52. Lynnerup N (1998) The Greenland Norse. Meddelelser om Grønland, Man & Society 24:149. 53. McGovern TH (1985) Contributions to the paleoeconomy of Norse Greenland. Acta Archaeologica

54: 73-122.

24

54. Arneborg J, Lynnerup N, Heinemeier J (2012) Human diet and subsistence patterns in Norse Greenland AD c.980–AD c.1450: Archaeological Interpretations. Journal of the North Atlantic. Special Volume 3:119–133.

55. Bichet V, Gauthier E, Massa C, Perren B (2014) Lake sediments as an archive of land use and environmental change in the Eastern Settlement, Southwestern Greenland. Journal of the North Atlantic. Special Volume 6:47-63.

56. Edwards KJ, Schofield JE, Mauquoy D (2008) High resolution paleoenvironmental and chronological investigations of Norse landnám at Tasiusaq, Eastern Settlement, Greenland. Quaternary Research 69:1–15.

57. Arneborg J (2008) The Norse settlements in Greenland. The Viking World, eds Brink S, Price, N (Routledge, London), pp 588-597.

58. Madsen CK (2014) Pastoral settlement, farming, and hierarchy in Norse Vatnahverfi, South Greenland. PhD dissertation (Univ. of Copenhagen, Copenhagen).

59. Roussell A (1941) Farms and churches in the medieval Norse settlements of Greenland. Meddelelser om Grønland 89:354.

60. Arneborg J (1991) The Roman Church in Norse Greenland. The Norse of the Norse Atlantic. Acta Archaeologica 61(1990):142-150.

61. Golding KA, Simpson IA, Schofield JE, Edwards KJ (2011) Norse–Inuit interaction and landscape change in Southern Greenland? A geochronological, pedological, and palynological investigation. Geoarchaeology: An International Journal 26(3):315–345.

62. Magerøy H (1993) Soga om Austmenn: Nordmenn som siglde til Island og Grønland i mellomalderen (Norske samlaget, Oslo).

63. Arge SV, Sveinbjarnardóttir G, Edwards KJ, Buckland PC (2005) Viking and Medieval settlement in the Faroes: People, place, and environment. Hum Ecol 33(5):597-620.

64. Mahler DL (1998) The stratigraphical cultural landscape. Outland Use in Preindustrial Europe, eds Andersson H, Ersgård L, Svensson E (Lund Studies in Medieval Archaeology 20, Institute of Archaeology, Univ. of Lund, Lund), pp 49-62.

65. Thorsteinsson A (2008) Land division, land rights, and landownership in the Faeroe Islands. Nordic Landscapes: Region and Belonging on the Northern Edge of Europe, eds Jones M, Olwig K (Univ. of Minnesota Press, Minneapolis, MN), pp 77-105).

66. Church M, et al. (2005) Puffins, pigs, cod and barley: Palaeoeconomy at Undir Junkarinsfløtti, Sandoy, Faroe Islands. Environmental Archaeology 10:179-197.

67. Brewington SD (2011) Fourth interim report on analysis of archaeofauna from Undir Junkarinsfløtti, Sandoy, Faroe Islands. Report No. 56. of the NORSEC Zooarchaeology Laboratory (Northern Science and Education Center, City Univ. of New York, New York, NY).

68. Brewington SD (in press) The key role of wild resources in the Viking-Age to Late-Norse palaeoeconomy of the Faroe Islands: The zooarchaeological evidence from Undir Junkarinsfløtti, Sandoy. Climates of Change: The Shifting Environments of Archaeology: Proceedings of the 44th Annual Chacmool Conference, eds Lacey S, Tremain C, Sawyer M (Univ. Calgary Press, Calgary).

69. Thomson AM, Simpson IA, Brown JL (2005) Sustainable rangeland grazing in Norse Faroe. Hum Ecol 33:737-761.

70. Stummann Hansen, S (2013) The North Atlantic context and beyond. Acta Archaeologica 84:207-210.

71. Wylie J (1987) The Faroe Islands: Interpretations of History (Univ. Press of Kentucky, Lexington). 72. Debes LJ (1676) Færoæ & Færoa reserata, that is, A Description of the Islands & Inhabitants of

Foeroe, Being Seventeen Islands Subject to the King of Denmark, Lying under 62 Deg. 10 Min. of North latitude : Wherein Several Secrets of Nature are Brought to Light, and Some Antiquities Hitherto Kept in Darkness Discovered (Flower-de-Luc, London).

25

73. Williamson K (1970) The Atlantic Islands: A Study of the Faeroe Life and Scene, 2nd ed (Routledge & Kegan Paul, London).

74. Brink S (2008) Law and society: Polities and legal customs in Viking Scandinavia. The Viking World, eds Brink S, Price N (Routledge, London), pp 23-31.

75. Joensen JP (1996) The fisheries of the Faroe Islands: An overview. The North Atlantic Fisheries, 1100-1976: National Perspectives on a Common Resource, eds Holm P, Starkey DJ, Thór JT (Studia Atlantica, Esbjerg, Denmark), pp 27-47.

76. Blake T, LeBlanc S, Minnis PE (1986) Changing settlement and population in the Mimbres Valley, SW New Mexico. Journal of Field Archaeology 13:439-464.

77. Hegmon M, et al. (2008) Social transformation and its costs in the prehistoric US Southwest. Am Anthropol 110(3):313-324.

78. Minnis PE (1985) Social Adaptation to Food Stress: A Prehistoric Southwestern Example. (Univ of Chicago Press, Chicago IL).

79. Spielmann KA, Nelson MC, Peeples MA, Ingram SE (2011 ) Sustainable small-scale agriculture in semi-arid environments. Ecol Soc 16(1):26.

80. Schollmeyer K (2009) Resource stress and settlement pattern change in the eastern Mimbres area, southwest New Mexico. PhD dissertation (Arizona State Univ., Tempe, AZ).

81. Cannon MD (2001) Large mammal resource depression and agricultural intensification: An empirical test in the Mimbres Valley, New Mexico. PhD dissertation (Univ. of Washington, Seattle, WA).

82. Sandor J, Gersper PL, Hawley JW (1990) Prehistoric agricultural terrace and soils in the Mimbres area, New Mexico. World Archaeol 22(1):70-86.

83. Hegmon M, Nelson MC (2007) In sync, but barely in touch: Relations between the Mimbres region and the Hohokam regional system. Hinterlands and Regional Dynamics in the Ancient Southwest, eds Sullivan AP, Bayman J (Univ. of Arizona Press, Tucson, AZ), pp 70-96.

84. Brody JJ (2004) Mimbres Painted Pottery, Revised Edition (School of American Research Press, Santa Fe, NM).

85. Gilman PA, Thompson M, Wyckoff KC (2014) Ritual change and the distant: Mesoamerican iconography, Scarlett Macaws, and great kivas in the Mimbres region of southwestern New Mexico. Am Antiq 79(1):90-107.

86. Bradfield M (1971) The Changing Pattern of Hopi Agriculture (Royal Anthropological Institute of Great Britain and Ireland, London).

87. Ford RI (1968) An ecological analysis involving the population of San Juan Pueblo, New Mexico. PhD dissertation (Univ. of Michigan, Ann Arbor, MI).

88. Forde CD (1931) Hopi agriculture and land ownership. J R Anthropol Inst 61:357-405. 89. Hegmon M, Brady JA, Nelson MC (2006) Variability in Classic Mimbres room suites: Implications for

household organization and social differences. Mimbres Society, eds Powell-Marti VS, Gilman PA (Univ. of Arizona Press, Tucson, AZ), pp 45-65.

90. Shafer H (1982) Classic Mimbres phase households and room use patterns. Kiva 48 (1-2):17-37. 91. Anyon R, LeBlanc SA (1984) The Galaz Ruin: a Prehistoric Mimbres Village in Southwestern New

Mexico (Univ. of New Mexico Press, Albuquerque, NM). 92. Shafer HJ (1999) The Mimbres Classic and Postclassic: A case for discontinuity. The Casas Grandes

World, eds Schaafsma CF, Riley CL (Univ. of Utah Press, Salt Lake City), pp 121-133. 93. Shafer H (2006) Extended families to corporate groups. Mimbres Society, eds Powell-Martí VS,

Gilman PA (Univ. of Arizona Press, Tucson, AZ), pp 15-31. 94. McAnany PA (1998) Ancestors and the Classic Maya built environment. Function and Meaning in

Classic Maya Architecture, ed Houston SD (Dumbarton Oaks, Washington, DC), pp 271-298.

26

95. McGuire RH, Saitta DJ (1996) Although they have petty captains, they obey them badly: The dialectics of prehispanic Western Pueblo social organization. Am Antiq 61(2):197-216.

96. Nelson MC (1986) Chipped stone analysis: Food selection and hunting behavior. Short-Term Sedentism in the American Southwest: The Mimbres Valley Salado, eds Nelson BA, LeBlanc SA (Univ. of New Mexico Press, Albuquerque, NM), pp. 141-176.

97. Schollmeyer KG, Driver JC (2013) Settlement patterns, source-sink dynamics, and artiodactyl hunting in the prehistoric U.S. Southwest. Journal of Archaeological Method and Theory. Special Volume 20:448-487.

98. Nelson MC (1999) Mimbres during the 12th Century: Abandonment, Continuity, and Reorganization (Univ. of Arizona Press, Tucson).

99. Schollmeyer KG, Russell WG, Nelson MC (2010) Excavations at two thirteenth-century villages west of Truth or Consequences. NewsMAC 4:2-13.

100. Nelson BA, Anyon R (1996) Fallow valleys: Asynchronous occupations in southwestern New Mexico. Kiva 61:275-294.

101. Kintigh KW, Glowacki DM, Huntley D (2004) Long-Term settlement history and the emergence of towns in the Zuni area. Am Antiq 68(4):432-456.

102. Kintigh KW (2007) Late prehistoric and protohistoric settlement systems in the Zuni area. Zuni Origins: Toward a New Synthesis of Southwestern Archaeology, eds Gregory DA, Wilcox DR (Univ. of Arizona Press, Tucson, AZ), pp 361-376.

103. Howell TL (2000) The Archaeology and ethnohistory of Oak Wash, Zuni Indian Reservation, New Mexico. Zuni Cultural Resource Enterprise Report No. 664 (Zuni Cultural Resource Enterprise, Pueblo of Zuni, NM).

104. Varien MD (1990) Excavations at three prehistoric sites along Pia Mesa Road, Zuni Indian Reservation, McKinley County, New Mexico. Zuni Archaeology Program Report No. 233 (Zuni Archaeology Program, Pueblo of Zuni, NM).

105. Duff A, Moss J, Windes T, Kantner J, Shackley S (2012) Patterning in procurement of obsidian in Chaco Canyon and Chaco-era communities in New Mexico as revealed by X-ray fluorescence. J Archaeol Sci 39(9):2995-3007.

106. Nelson MC, Kintigh KW, Abbott DR, Anderies JM (2010) The cross-scale interplay between social and biophysical context and the vulnerability of irrigation-dependent societies: Archaeology's long-term perspective. Ecol Soc 15 (3): 31.

107. Spielmann KA (1989) Colonists, hunters and farmers: Plains-Pueblo interaction in the seventeenth century. Columbian Consequences, Volume 1, ed Thomas DH (Smithsonian Institution Press, Washington, DC), pp 101-113.

108. Spielmann KA, Clark T, Hawkey D, Rainey K, Fish SK (2009) “…being weary, they had rebelled”: Pueblo subsistence and labor under Spanish colonialism. J Anthropol Archaeol 28:102-125.

109. Chamberlin M (2008) Evaluating the cultural origins of complexity in the Ancestral Pueblo world. PhD dissertation (Arizona State Univ., Tempe, AZ).

110. Rautman AE (1995) Final report on the excavation of Pueblo de la Mesa (US Dept. of Agriculture, US Forest Service, Cibola National Forest, Albuquerque, NM).

111. Rautman AE (2013) Final report on the excavation of LA-9032, Frank’s Pueblo, Socorro County, New Mexico (State of New Mexico Department of Cultural Affairs, Historic Preservation Division, Santa Fe, NM).

112. Hill JB, et al. (2004) Prehistoric demography in the Southwest: Migration, coalescence, and Hohokam population decline. Am Antiq 69(4):689–716.

113. Kwiatkowski SM (2003) Evidence for subsistence problems. Centuries of Decline during the Hohokam Classic Period at Pueblo Grande, ed Abbott DR (Univ. of Arizona Press, Tucson, AZ), pp 48-69.

27

114. Dean R (2007) The Lagomorph Index: Rethinking rabbit ratios in Hohokam sites. Kiva 73(1):7-30. 115. James SR (2003) Hunting and fishing patterns leading to resource depletion. Centuries of Decline

during the Hohokam Classic Period at Pueblo Grande, ed Abbott DR (Univ. of Arizona Press, Tucson, AZ), pp 70–81.

116. Szuter CR (1989) Hunting by Prehistoric Horticulturalists in the American Southwest (Univ. of Arizona Press, Tucson).

117. Abbott DR, Smith AM, Gallaga E (2007) Ballcourts and ceramics: The case for Hohokam marketplaces in the Arizona desert. Am Antiq 72(3):461-484.

118. Anderies JM (2006) Robustness, institutions, and large-scale change in social-ecological systems: The Hohokam of the Phoenix Basin. Journal of Institutional Economics 2(02):133-155.

119. Abbott DR, Ingram SE, Kober BG (2006) Hohokam exchange and early Classic period organization in central Arizona: Focal villages or linear communities? Journal of Field Archaeology 31(3):285-305.

120. Graybill DA, Nials FL (1989) Aspects of climate, streamflow and geomorphology affecting irrigation systems in the Salt River Valley. The 1982-1984 Excavations at Las Colinas: Environment and Subsistence, eds Graybill DA, et al. (Cultural Resource Management Division, Arizona State Museum, University of Arizona, Tucson), pp 5-23.

121. Rice G (1998) War and water: an ecological perspective on Hohokam irrigation. Kiva 63(3):263-301. 122. Henderson TK, Clark TC (2004). Changing perspectives: Considerations of past agricultural use of

the Salt River floodplain. Hohokam Farming on the Salt River Floodplain: Refining Models and Analytical Methods, ed Henderson TK (Anthropological Papers No. 43, Center for Desert Archaeology, Tucson, AZ), pp 167-187.

123. Howard JB (2006) Hohokam irrigation communities: A study of internal structure, external relationships and sociopolitical complexity. PhD dissertation (Arizona State Univ., Tempe, AZ).

124. Woodson MK (2010) The social organization of Hohokam irrigation in the Middle Gila River Valley, Arizona. PhD dissertation (Arizona State Univ., Tempe, AZ).

125. Means TH (1901) Soil Survey in Salt River Valley, Arizona. Agricultural Experiment Station Bulletin 40 (Univ. of Arizona, Tucson, AZ).

28


Supporting Information #3: Conditions after climate challenge

Margaret C. Nelson, et al.2

Supporting Information #3 includes the archaeological and historical evidence used to support codes

describing the nature of change following each of the thirteen climate challenges in the seven regions in

this study. Climate challenges are defined and described further in SI#1; regions are delineated and

illustrated in SI#2. The codes below are based on change occurring during and immediately after the

period marked as a climate challenge. Coding was done by all co-authors in face-to-face discussion.

Social/demographic changes are coded with supporting evidence in Table S3-1. The codes for the

nature of social changes associated with climate challenges are 1 = little change, 2 = substantial change,

and 3 = transformative change.

1 - Challenges/cases coded as 1 represent contexts where there were no major changes in

population levels and no loss of institutions, but may include cases marked by minor changes in the

organization of population or economic practices. Cases coded 1 are characterized by substantial

continuity.

2 - Challenges/cases coded as 2 represent cases where there were major changes in the

organization or distribution of populations or major changes in institutional reach or function, but also

substantial evidence for continuity.

3 - Challenges/cases coded as 3 represent major instances of change marked by major

population loss (depopulations), dramatic changes in social organization, and the loss of institutions or

infrastructure. Many cases marked 3 represent the terminal phases of occupation of regions by a given

social group or most of that social group.

Codes and supporting archaeological and historical evidence are presented by the climate challenges

identified within each case. Each code characterizes changes immediately following the date of the

initiation of the climate challenge.

2 All authors to the article as listed on the cover page of this SI document.

29

Table S3-1. Social/demographic changes following the climate challenge.

CASE

CLIMATE CHALLENGE START CODE EVIDENCE

Iceland 1 1257 2

End of commonwealth; dramatic social changes in laws and independence; previous legal and social relations abolished (1). Legal structure remains same and civil war continues (1).

Iceland 2 1310 1 Continuation of the new royal administration (1).

Iceland 3 1608 1 Many changes in administration and global relations; integrated into Danish state (1) but locally similar to previous centuries.

Greenland 1 1257 2

Depopulation of the marginal regions and decline in agro-pastoralism. Livestock numbers were reduced and flocks of sheep and goats increased to the exclusion of cattle (2-5). Dependence on outer fjord marine resources (migrating seals) increased. Settlement contracted to the inner and mid fjord regions and the authority and central functions of society including economic power was gathered on fewer hands, witnessed by closed down churches and the building of fewer larger and more monumental structures (2). Norse Greenlanders were now under Norwegian authority, which may have impacted the local authority. The influence of the Norwegian bishops is uncertain (6).

Greenland 2 1310 2

Change in food mix that had impact on the food base and on organization of food getting that ultimately affected the outcome of Norse occupation (2, 5).

Greenland 3 1421 3

Abandonment of the Norse settlement was definitive around 1450. The northern Western settlement was abandoned a few generations before the southern Eastern Settlement (7, 8).

Faroes 1257 2

There may have been a further restriction of access to resources for the poor (as exemplified by the Sheep Letter of 1298); this was part of a long period of social change (much of which was driven by Norwegian forces (9, 10).

Mimbres 1 1127 3

Major population decline and social reorganization evident in settlement organization and domestic architecture (11, 12).

Mimbres 2 1273 1

No evidence of regional changes in settlement organization or social relations. From 1150 on, the region was sparsely populated with considerable diversity in pottery, architectural, and other material forms and styles (11, 13).

Zuni 1133 1

Region-wide change with limited local impact. This period saw the end of the Chaco phenomenon, which was a major force in the earlier century. In some communities in the Zuni area, things kept going right on as they had previously including the continued occupation of great house/great kiva communities. In a few other places, people began to build post-Chacoan architectural complexes with a decidedly more local flavor but there was apparently a lot of continuity in community organization (14).

30

Salinas 1335 2

End of settlement reorganization from jacals to masonry pueblos. Documented burning is interpreted as evidence for conflict in the mid 1300s that immediately followed the “reorganization” transformation (15).

Hohokam 1 1338 2

Beginning of substantial depopulation of the Phoenix Basin. The abandonment of the Phoenix Basin (see below) was preceded by a slow demographic decline in the late Classic Period from the AD 1300s until abandonment of the region in the mid AD 1400s (16, 17).

Hohokam 2 1436 3

End of the Classic Period, which saw the depopulation of the entire Phoenix Basin and the abandonment of the massive irrigation canal system (16, 18).

In addition to social/demographic change, changes in food conditions following the climate challenge

were identified. Food conditions include actual suffering of hunger but may also include difficulty

getting to food resources (as in Greenland). Changes in food shortage were coded from 1 to 3. 1 = little

change, 2 = more shortage for some or a small decline for all, and 3 = substantial decline for all.

Table S3 2. Declines in food shortage following climate challenges

CASE

CLIMATE CHALLENGE START CODE EVIDENCE

Iceland 1 1257 2

Starvation increased as a result of poor weather (19). As noted by Ogilvie (20) the Icelandic Annals for 1287 state that: “At this time, many severe winters came at once, and following them, people died of hunger.” Extreme bone processing of harp seals suggests nutritional stress (follow Outram's study standards [21]) from the ca. 1250-1300 deposit at Hofstadir (22).

Iceland 2 1310 2

Increasing use of seals and marine fish at inland settlements suggests use of secondary resources (22). Continued references to famine and shortages but evidence of substantial population up to impact of Black Death in early 15th c. (23).

Iceland 3 1608 2 Recurring famines as always, but increasingly bad for lower class (1).

Greenland 1 1257 1 No evidence for shortage.

Greenland 2 1310 1 No evidence for shortage.

Greenland 3 1421 3

Seals had kept people alive, and as long as the landowners controlled the seal hunt the seals also kept the social system alive, at least on the face of it. The seals also masked serious structural problems, which perhaps are expressed in the social equalization of diet. Not even the elite farms were profitable anymore. This climate challenge made sealing difficult contributing, among many factors, to the end of Norse settlement. (8).

Faroes 1257 1 Evidence is extremely limited, but there is no evidence of food

31

shortfalls.

Mimbres 1 1127 1 No skeletal evidence of food shortage (24); no substantial changes in diet across this century (11, 25).

Mimbres 2 1273 1 No evidence of reduction in available food.

Zuni 1133 1

No evidence for a change in food availability across the transition considered here; similar frequencies and diversities of foods have been noted at sites occupied before and after the transition (e.g., 26-28).

Salinas 1335 2

The hostile landscape and burning of stored food, including corn (28), would have created some food shortfall or certainly a food challenge.

Hohokam 1 1338 2

Canal system declining in extent (30, 31), and streamflow was variable (32). Archaeological research has shown that health was declining at some places during the Classic Period (33), although McClelland and Lincoln-Babb, in a conference presentation, claim that poor health was not widespread and originally may have been overstated. Nevertheless some poor nutrition is evident (16).

Hohokam 2 1436 3 Abandonment of the canal massive canal system at this time (16, 18) would have prompted considerable food shortage locally.

References

1. Karlsson G (2000) Iceland’s 1100 years: The History of a Marginal Society (C. Hurst, London). 2. Arneborg J, Lynnerup N, Heinemeier J (2012) Human diet and subsistence patterns in Norse

Greenland AD c.980–AD c.1450: Archaeological Interpretations. Journal of the North Atlantic. Special Volume 3:119–133.

3. Bichet V, Gauthier E, Massa C, Perren B (2014) Lake sediments as an archive of land use and environmental change in the Eastern Settlement, Southwestern Greenland. Journal of the North Atlantic. Special Volume 6:47-63.

4. Madsen CK (2014) Pastoral settlement, farming, and hierarchy in Norse Vatnahverfi, South Greenland. PhD dissertation (Univ. of Copenhagen, Copenhagen).

5. McGovern TH (1985) Contributions to the paleoeconomy of Norse Greenland. Acta Archaeologica 54: 73-122.

6. Arneborg J (1991) The Roman Church in Norse Greenland. The Norse of the Norse Atlantic. Acta Archaeologica 61(1990):142-150.

7. Arneborg J (2008) The Norse settlements in Greenland. The Viking World, eds Brink S, Price, N (Routledge, London), pp 588-597.

8. Dugmore AJ, et al. (2012) Cultural adaptation, compounding vulnerabilities, and conjunctures in Norse Greenland. Proc Natl Acad Sci U S A 109(10):3011-3016.

9. Mahler DL (1998) The stratigraphical cultural landscape. Outland Use in Preindustrial Europe, eds Andersson H, Ersgård L, Svensson E (Lund Studies in Medieval Archaeology 20, Institute of Archaeology, Univ. of Lund, Lund), pp 49-62.

10. Wylie J (1987) The Faroe Islands: Interpretations of History (Univ. Press of Kentucky, Lexington). 11. Nelson MC (1999) Mimbres during the 12th Century: Abandonment, Continuity, and

Reorganization (Univ. of Arizona Press, Tucson).

32

12. Nelson MC, Hegmon M, Kulow S, Schollmeyer KG (2006) Archaeological and ecological perspectives on reorganization: A case study from the Mimbres region of the US Southwest. Am Antiq 71(3):403-432.

13. Schollmeyer KG, Russell WG, Nelson MC (2010) Excavations at two thirteenth-century villages west of Truth or Consequences. NewsMAC 4:2-13.

14. Kintigh KW, Howell TL, Duff AI (1996) Post-Chacoan social integration at the Hinkson Site, New Mexico. Kiva 61:257-274.

15. Rautman AE (2013) Final report on the excavation of LA-9032, Frank’s Pueblo, Socorro County, New Mexico (State of New Mexico Department of Cultural Affairs, Historic Preservation Division, Santa Fe, NM).

16. Abbott DR, ed (2003) Centuries of decline during the Hohokam Classic period at Pueblo Grande. (Univ. of Arizona Press, Tucson).

17. Ingram SE (2010) Human vulnerability to climatic dry periods in the prehistoric U.S. Southwest. PhD dissertation (Arizona State Univ., Tempe, AZ).

18. Wilcox DR, Masse WB eds (1981) The Protohistoric period in the North American Southwest, AD 1450-1700. Anthropological Papers No. 24 (Arizona State University, Tempe, AZ).

19. Ogilvie AEJ, Woollett JM, Smiarowski K (2009) Seals and sea ice in medieval Greenland. Journal of the North Atlantic 2:60-80.

20. Ogilvie AEJ (1991) Climatic changes in Iceland A.D. c. 1500 to 1598. The Norse of the North Atlantic, ed Bigelow GF (Acta Archaeologica 61(1900), Munksgaard, Copenhagen, Denmark), pp 233–251.

21. Outram AK (1999) A comparison of Paleo-Eskimo and Medieval Norse bone fat exploitation in Western Greenland. Arctic Anthropol 36(1-2):103-117.

22. McGovern TH, Harrison R, Smiarowski K (2014) Sorting sheep & goats in Medieval Iceland and Greenland: Local subsistence or world system? Long-Term Human Ecodynamics in the North Atlantic: An Archaeological Study, eds Harrison R, Maher R (Lexington, Lanham, MD), pp 153-175.

23. Eggertsson T (1998) Sources of risk, institutions for survival, and a game against nature in premodern Iceland. Explor Econ Hist 35:1–30.

24. Lippmeier HS (1991) A preliminary assessment of the relative health status between classic and post-classic populations of southwestern New Mexico. MA thesis (State Univ. of New York, Buffalo, NY).

25. Hegmon M, Nelson MC, Schollmeyer KG, Elliott M, Diehl M (2006) Agriculture, mobility, and human impact in the Mimbres region of the US Southwest. Managing Archaeological Data and Databases: Essays in Honor of Sylvia W. Gaines, eds Hantman JL, Most R (Anthropological Research Paper #55, Arizona State University, Tempe, AZ), pp 105-119.

26. Howell TL (2000) The Archaeology and ethnohistory of Oak Wash, Zuni Indian Reservation, New Mexico. Zuni Cultural Resource Enterprise Report No. 664 (Zuni Cultural Resource Enterprise, Pueblo of Zuni, NM).

27. Varien MD (1990) Excavations at three prehistoric sites along Pia Mesa Road, Zuni Indian Reservation, McKinley County, New Mexico. Zuni Archaeology Program Report No. 233 (Zuni Archaeology Program, Pueblo of Zuni, NM).

28. Zier CJ (1976) Excavations near Zuni, New Mexico: 1973. Museum of Northern Arizona Research Paper No. 2 (Museum of Northern Arizona, Flagstaff, AZ).

29. Rautman AE (1995) Final report on the excavation of Pueblo de la Mesa (US Dept. of Agriculture, US Forest Service, Cibola National Forest, Albuquerque, NM).

30. Howard JB (2006) Hohokam irrigation communities: A study of internal structure, external relationships and sociopolitical complexity. PhD dissertation (Arizona State Univ., Tempe, AZ).

33

31. Woodson MK (2010) The social organization of Hohokam irrigation in the Middle Gila River Valley, Arizona. PhD dissertation (Arizona State Univ., Tempe, AZ).

32. Graybill DA, Gregory DA, Funkhouser GS, Nials FL (2006) Long-term streamflow reconstructions, river channel morphology, and aboriginal irrigation systems along the Salt and Gila Rivers. Environmental Change and Human Adaptation in the Ancient American Southwest, eds Doyel DE, Dean JS (Univ. of Utah Press, Salt Lake City), pp 69–123.

33. Sheridan SG (2003). Childhood health as an indicator of biological stress. Centuries of Decline during the Classic Period Hohokam at Pueblo Grande, ed Abbott DR (Univ. of Arizona Press, Tucson, AZ), pp 82-106.

Documents

Climate challenges, vulnerabilities, and food security · 2016-01-01 · security, one of the seven human securities identified by a United Nations Human Development Report (15) (see