A framework for the science contribution in climate adaptation: Experiences from science-policy processes in the Andes

A

framework

for

the

science

contribution

in

climate

adaptation:

Experiences

from

science-policy

processes

in

the

Andes

Christian

Huggel

a,

*

,

Marlene

Scheel

a

,

Franziska

Albrecht

b,c

,

Norina

Andres

d

,

Pierluigi

Calanca

e

,

Christine

Jurt

a,e

,

Nikolay

Khabarov

b

,

Daniel

Mira-Salama

f

,

Mario

Rohrer

g

,

Nadine

Salzmann

a,h

,

Yamina

Silva

i

,

Elizabeth

Silvestre

j

,

Luis

Vicun˜

a

a

,

Massimiliano

Zappa

d

a_{DepartmentofGeography,UniversityofZurich,8057Zurich,Switzerland}

b_{InternationalInstituteforAppliedSystemsAnalysis,2361Laxenburg,Austria}

c

DepartmentofGeographyandRegionalResearch,UniversityofVienna,1010Vienna,Austria

d_{SwissFederalInstituteofForest,SnowandLandscapeResearch,8903Birmensdorf,Switzerland}

e_{Agroscope,InstituteforSustainabilitySciencesISS,8046Zurich,Switzerland}

f_{EnvironmentandNaturalResourcesGlobalPractice,TheWorldBank,Washington,DC20433,UnitedStates}

g_{MeteodatGmnbH,8005Zurich,Switzerland}

h

DepartmentofGeosciences,UniversityofFribourg,1700Fribourg,Switzerland

i

GeophysicalInstituteofPeru,Lima,Peru

j_{INCLIMA,PuebloLibre,Lima,Peru}

Assignificantimpactsofclimatechangeareincreasinglyconsideredunavoidable,adaptationhas become a policypriority.Itis generally agreedthatscienceis importantfor the adaptationprocess

butspecificguidanceonhowandtowhatdegreescienceshouldcontributeandbeembeddedin

thisprocessisstilllimitedwhichisatoddswiththehighdemandforsciencecontributionsto

climateadaptationbyinternationalorganizations,nationalgovernmentsandothers.Herewe

presentandanalyzeexperiencesfromthetropicalAndesbasedonarecentscience-policyprocess

onthenationalandsupra-nationalgovernmentlevel.Duringthisprocessaframeworkforthe

sciencecontributioninclimateadaptationhasbeendeveloped;itconsistsofthreestages,

including(1) the framing and problemdefinition, (2)thescientificassessmentofclimate,impacts, vulnerabilitiesandrisks,and(3)theevaluationofadaptationoptionsandtheirimplementation.A

largeamountofmethodshasbeenanalyzedforstage(2),andanumberofmajorclimate

adaptationprojectsintheregionassessedfor(3).Ourstudyunderlinestheimportanceofjoint

problemframingamongvariousscientificandnon-scientificactors,definitionof

socio-environ-mentalsystems,timeframes,andamoreintenseinteractionofsocialandphysicalclimateand

impactsciences.Scientifically,thescarcityofenvironmental,socialandeconomicdata inregions

liketheAndescontinuetorepresentalimitationtoadaptation,andfurtherinvestmentsinto

coordinatedsocio-environmentalmonitoring,dataavailabilityandsharingareessential.

Published in (QYLURQPHQWDO6FLHQFH 3ROLF\±

which should be cited to refer to this work.

1.

Introduction

Aseffectivemitigationofclimatechangehasturnedouttobea

major political challenge, and the objective of containing

warmingtobelow+28Cinrelationtopreindustriallevelsis

increasinglydifficulttoachieve(Stocker,2013),adaptationto

bothnegativeandpositiveeffectsofwarmingisunavoidable

andhasbecomeapolicypriority.Infact,overthepastyears,

increasingattentionhasbeenpaidtoadaptation,reflectedin

thenegotiationsoftheUnitedNationsFrameworkConvention

on Climate Change (UNFCCC) (Khan and Roberts, 2013).

Adaptationisrelevantbothforthedevelopinganddeveloped

world, and countries have developed adaptation strategies

and beguntoplan and implement adaptation measuresin

differentsectors,suchaswaterresources,naturalhazardsor

agriculture(Biesbroeketal.,2010;BotzenandVanDenBergh,

2008;Poussinetal.,2012).

It has been recognized that adaptation is a multi-scale

processthatneedstoconsidernotonlydifferentsourcesof

knowledgebutalsosocietalandculturalvalues,objectivesand

risk perceptions of different involved actors (IPCC, 2014).

Adaptationisplace-andcontext-specific,withlocal

govern-ments and communities playinga keyroleinthe process.

How,whenandtowhatdegreescienceshouldinteractandbe

partoftheadaptationprocessisanongoingdiscussionand

hasalsobeenanalyzedforotherenvironmentalfields(Moser

andEkstrom,2010;Paavola,2008;Pohlet al.,2010).Science

playsaparticularroleintheknowledgeproductionprocess

whereothersourcesofknowledgesuchaslocaland

indige-nous knowledge may also be of substantial relevance,

depending on the context (Cash et al., 2003; Ford et al.,

2012; Valdivia et al., 2010; Vogel et al., 2007). Research on

science-policyinteractionshasanalyzedandproposed

differ-ent models. A supply driven model, also termed ‘science

push’,foreseesthatscienceproductionisdrivenbythepursuit

of knowledge with limited applicability to solution of

problems, and thus often limited use for decision-making

(Bielaketal.,2008;Cashetal.,2006;DillingandLemos,2011).In

a demand driven, ‘policy pull’ model the production of

scientificknowledgeiscommissionedbypolicyinsearchof

practical solutions, but various experiences show that the

informationaskedformaynotbecoherentwiththescientific

perspective (Sarewitz and Pielke, 2007). A third approach

combinesthe‘sciencepush’and‘policypull’modelsintoa

co-production,orjointproductionofknowledgewhichimpliesan

iterativecommunicationbetweenscienceandpolicy(Dilling

and Lemos, 2011; Lemos and Morehouse, 2005). Based on

empirical evidence, there is increasingagreement that the

approachofjointproductionofknowledgeisamorefeasible

and successful model to achieve environmental solutions

(Heggeretal.,2012).Thecallisoutforadaptationresearchto

analyzehowdecisionsaremade,and howtofacilitateand

improvethepolicyimplementationunderconstraintsoftime,

information,capacityandresources(Smithetal.,2009).

Inpractice,todate,weobservedastrongdemandbypolicy

for science to develop and contribute primarily toward a

scientific basis for adaptation projects in fields such as

physical climate science, climate impacts, vulnerabilities

and risks.Thescientificbasisistherebyunderstoodas the

dataandinformationthatisgeneratedbyscienceregarding

thenaturalandhumansystems,typicallyincluding

observa-tionsontherecentpast,andprojectionsandscenariosofthe

future. An analysis of the actual adaptation process that

includesadaptation planning and managementunder

con-siderationofdifferentlevelsofgovernanceandlocalactors

hasbeenlessoftenaskedforandconductedbutis

neverthe-lessanessentialelement(Smithetal.,2009).

Thisstudy focuses on the scientific contributionto the

adaptation process, and how the scientific knowledge is

embeddedin thecontext ofthe adaptation process witha

focus on the tropicalAndes. As we will show here,it has

become clear that a framework for the production and

exchangeofscientificknowledgewithinthemultiple

dimen-sionsofadaptationplanningisanimportantneed.Recently

someframeworkshave indeedbeen proposed(seea

corre-spondingassessmentin(IPCC,2014),althoughsomeofthem

maylackspecificityintermsofthematicandmethodological

guidanceandregionalorlocalcontext.Severalinternational

organizationsanddevelopmentagencies,aswellastechnical

andscientificinstitutionsinfacthavecalledfor

methodologi-calguidance.ArecentinitiativeinvolvingtheWorldBank,the

AndeanCommunityofNations,theSwissFederalAgencyfor

the Environment, the Swiss Agency for Development and

Cooperation,andascientificconsortiumofSwissandAustria

basedresearchinstitutionsapproachedtheissue.Thefocusof

thestudywasonthetropicalAndesregion,yetwithaneyeon

apossibleapplicationoftheframeworkandmethodologyin

othermountainregions.Theselectionofthethematicscope

andscientificdisciplineswasbasedontheirimportance for

the Andes region and defined by the involved actors.

Accordingly,thefocuswasonclimatology,waterresources,

agricultureand food security, ecosystems, natural hazards

andrisks.Social,cultural,economicandpoliticalaspectswere

consideredascross-runningissuesimplyingahighly

inter-disciplinaryapproach.Itisrecognizedthatthethematicscope

defined here cannot encompass all potentially relevant

aspectsbutathematicconcentrationwasregardednecessary

forthesakeoffeasibility.

Theobjectivesofthispaperare:(i)developingaframework

forthesciencecontributionintheadaptationprocessinthe

Andescontext;(ii)providingevidenceandanalyzing

experi-ences for the different stages of the framework; and (iii)

putting the Andes experiences and framework in a larger

climateadaptationcontext.Theseobjectivesimplyacertain

hybridcharacterofthispaper,bypresentingaframeworkand

discussing its implementation, but also including review

elementsonthestateofmethodsindifferentthematicfields

intheAndesregionandpartlybeyondtheregion.

Accordingly, the paper is organized as follows: we first

introduce the Andes region and the policy context within

which the framework was developed (Section 2). We then

describe the framework, along with its different stages,

including the framing and problem definition (Section 3),

thescientificassessmentofclimate,impacts,vulnerabilities

andrisks(Section4),andtheevaluationofadaptationoptions

andtheirimplementation(Section5).ForSections4and5we

buildon,and reviewexistingexperiencesfromscienceand

adaptationprojectsintheAndesregion,andanalyzetheuseof

scientificinformationintheseprojects.Weemphasizethatwe

concentrate on the science contribution in the adaptation

process;othercomponentsofadaptationmaybeequallyor

moreimportantbutarenotamainfocushere.Assuchthe

implementation of adaptation is beyond the scope of this

study but we include it in the framework because of its

fundamentalimportanceintheadaptationprocessandcycle.

In Section 6 we critically analyze the framework and the

relatedexperiencesinclimateadaptationintheAndes,and

identifykeygapsandneeds,andeventuallyputthisinalarger

(global) context.We are wellaware that all three subjects

addressed inSections3–5would meritaseparatestudy,at

least.However,weareconvincedthatthereisanaddedvalue

of integrating the complete process into one study, even

thoughsimplificationswillbenecessary.

2.

Study

region

and

science-policy

context

WeprimarilyfocusontropicalAndeanmountainand

high-mountain regions and exclude coastal or tropical forest

regionswhichalsoformpartofalltropicalAndeancountries

andaresimilarlysubjecttoadaptationinterests(Fig.1).Inthe

tropicalAndeanmountainsareas,theexposureand

vulnera-bilityofpeople toclimaticchange isgenerallyhigh dueto

often extremeconditions such ashigh climaticvariability,

steeptopography,remotesettingandhighlevels ofpoverty

(CAN,2008;Magrinetal.,2014;Salzmannetal.,2009).Rural

mountain communities in the Andes are in fact often

characterized by limited livelihood options and adaptive

capacityisconfinedby limitedinformation,pooraccess to

services and often inequitable access to productive assets

(GentleandMaraseni,2012;Lynch,2012;McDowellandHess, 2012;Sietzetal.,2012;YoungandLipton,2006).Furthermore,

dueto theirmore remotelocation mountain regions often

disposeofonlysparseclimaticandenvironmentalscientific

observations,adeficiencythatisparticularlypronouncedin

developing countries. These are important limitations for

elaboratingascientificbasisforclimatechange adaptation

planningandimplementation(Salzmannetal.,2009,2014).

ThisstudywasdevelopedinthecontextoftheAdaptation

totheImpactsofRapidGlacierRetreatintheTropicalAndes

project(PRAA)andinvolvedthefourcountriesoftheAndean

CommunityofNations(CAN),i.e.Bolivia,Colombia,Ecuador

andPeru.Thegoverningandexecutingpoliticalbodyofthe

projectwastheGeneralSecretariatoftheCAN.IntheCAN,any

regionaldecisionneedstoconciliateinterestsandexperiences

fromdifferentactorsandrealities.Thisimposeschallenges

andmightrequiretheuseofconsensus-buildingapproaches

andtools.FurtheractorsweretheMinistriesofEnvironment

ofthefourAndeannationalgovernments,plusanumberof

technical national government institutions such as the

nationalmeteorologicalandhydrologicalservices,orthecivil

defenseagencies.FromEuropeanumberofSwissandAustria

based universities and research institutions joined the

process. TheWorld Bank played a supervision roleof the

project,andactedastheimplementingagencyfortheGlobal

EnvironmentalFacility(GEF),whichprovidedtheGrantfunds

fortheproject.

3.

Framing

and

scoping

process

Aframeworkisoneofthecriticalelementsinaprocesswhere

scienceinteractswithpolicyandsociety(Dewulf,2013).Inour

casetheframeworkwasdevelopedduringanumberof

face-to-faceworkshopstakingplaceovermorethanoneyear.Fig.2

displays the result of this scoping exercise visualizing

differentstepsoftheadaptationprocessandtheembedment

ofthescientificcontributioninthecontextofdifferentactors,

includinglocalcommunities,governmentsandprivatesector,

withessentialsocial,cultural,economic, politicaland legal

dimensions.

In our framework the framing, scoping and problem

definition,andthedefinitionofthepertinent

socio-environ-mentalsystemsrepresentthefirststage.Weidentifiedthe

following questions that should be addressed at the first

stage:

 Whatsocio-environmentalsystemsareatstakeandhoware

theydefined?

 Which actors (e.g. local communities, institutions) are

involvedorrelevant?

 What is the social, cultural, political and environmental

context ofthe actors (historically and at present)? What

Fig.1–MapofthetropicalAndesregionandcountries,

indicatingthelocationoftheclimateadaptationprojects

analyzedinTable1:(1)PRAA;(2)PACC;(3)Proyecto

Glaciares;(4)INAP;(5)MacizoColombiano;(6)PPCR;(7)

IMACC.Backgroundimageistheprecipitationclimatology

forthemonthofJune,basedonTRMM3B43V7datafrom

1998to2013.

changescanbecomeimportantinthefuture,whataretheir consequences?

 Aretheexpectationsbythedifferentactorsinlinewiththe

scientificstate-of-the-art, in terms of methods,available

data,andtimeframe?

Studies of the dynamic social, political, cultural and

economiccontext,includingperceptionstudiescanidentify

socialandculturalaspectsoflocalenvironmentalchangeand

risks (Adger et al., 2013; Carey et al., 2012a,b) that are

important for the framing process. The objectives among

theactorsinvolvedinclimateadaptation,infact,cangreatly

diverge(Heggeretal.,2012;WeichselgartnerandKasperson,

2010). The definition of objectives and problem framing

thereforemustnecessarilybeajointprocessthatinvolvesa

rangeofstakeholdersandactors,wheretheinclusionoflocal

peopleandinstitutionshasbeenshowntoenhance

adapta-tionimplementationandreducesocialconflicts(Dewulf,2013;

Lynch,2012).

Theframingprocess (firststage) mayoverlaptime-wise

withthenext,second stagewhichconsistsofthescientific

assessments.Thislevelisexpectedtoproducetherequired

understanding of the socio-environmental systems, as

de-fined in the framing process, for past, present and future

conditions(seeSection4formoredetails).

Thethirdstagewhichmayalsooverlapwiththescientific

assessmentincludestheevaluationofadaptationoptionsand

eventually decisions on adaptation measures. The close

interactionbetweenvariousactorsgeneratedatthefirststage

ofthe process(framing,problemdefinition)isfacilitatinga

morecoherentandconcerted,andthususeful,processneeded

forthe developmentofadaptation strategiesand measures

(BrownandWilby,2012;Huggeletal.,2012;McNie,2007).The

open circle in Fig. 2 indicates the iterative nature of this

processwhich represents animportant finding frommany

adaptationexperiencesintheAndesandotherpartsofthe

world,asalsoemphasizedintherecentIPCC5thAssessment

Report(IPCC,2014).

Fig.2alsoincludesanindicativetimeframeofthedifferent

stepsoftheprocess.Thisaspecthasoftenbeenneglectedor

underestimatedinthepast,andexperiences(seealsoTable1)

showthatthedifferenttimeframesoftheinvolvedactorscan

be a major barrier to a successful science contribution in

adaptation(seeSection5).

4.

Scientific

assessments

Inthissectionwefocusonthescientificassessmentasamain

componentofthesciencecontributionwithintheframework

process(Fig.2).Wedistinguishtwomainlevelsthatinclude

the thematic studies and can broadly be sketched as

observationbased (level 1) and projection(scenario) based

(level2)(Fig.3).Theoutcomeofthefirst,observationbased

level,essentiallyresultsinanimprovedunderstandingofthe

relevantsystems,theirvulnerabilities,sensitivities,resilience

andadaptivecapacities.Theidentificationofcritical

knowl-edgegapsanduncertaintiesissimilarlyimportant.

Inthe second, projectionbased level,basic information

comesfromarangeofscenarios forthefuture.Howthese

scenarios are generated (e.g. by participatory methods,

analysisofdriversofchange,etc.)(Kriegleretal.,2012;O’Neill

et al., 2014) should be consistent with the definition of

objectivesandproblemframing.Thescenariooutcomeneeds

tobetranslatedor propagatedinto therelevant topicsand

disciplines.Again,whichmethods areusedforthisprocess

essentiallydependsontheinitialproblemframing,available

data and information, resources available etc. Adaptation

involves essential social questions, which should be

ade-quatelyconsidered inthe scientificassessment and

knowl-edgeproduction.Hereweenvisionthehumandimensionas

runningacrossthedifferentthematicaspects,eventhoughin

the structure that follows social, cultural, economic and

political aspects are addressed in a separate section. We

therebyemphasizetheimportanceofputtingthe scientific

resultsandmodelsintherespectivesocial,cultural,economic

andpoliticaldynamiccontext.

Foreachthematicaspect(seealsoFig.3)wereviewand

analyze the state of methods and results that has been

producedsofarinthetropicalAndes.Thisanalysishasbeen

doneduringtheaforementioned workshopsand additional

onlinemeetingsinvolvingAndesandEuropebasedexperts.As

aresultfivedraftdocumentswereproduced,criticallyrevised,

discussed, adapted,complementedand brought inits final

formthateventuallywasapprovedbyallfourgovernmentsof

Fig.2–Visualizationoftheframeworkforaclimate

adaptationprocesswithsciencecontribution.Threemain

stagesaredistinguishedandindicatedasboxes,where

bluedenotesprocesseswithastrongpolicyand

stakeholdercomponent,andgreenwithastrongerfocus

onscience.Thedashedarrowindicatestheiterative

natureoftheadaptationprocess.Arrowsattheright

provideanapproximatedurationoftherespectivestageof

theframework.(Forinterpretationofthereferencesto

colorinthisfigurelegend,thereaderisreferredtotheweb

versionofthisarticle.)

Table 1 – Analysis of science-policy interaction and context of selected major climate adaptation projects in the tropical Andes with an explicit science component.

Project name and duration

Main funding/policy institutions

Countries and sectors involved

Objectives Scientific input

(type and methods); interdisciplinary scope

Joint knowledge production process: problem framing and

system definition Use of scientific information for adaptation measures PRAAa 2008–2014

World Bank, Andean Community of Nations, national governments

Bolivia, Ecuador, Peru, (Colombia)

Glaciers, water resources, ecosystems, agriculture

To contribute to strength-ening the resilience of local ecosystems and economies to the impacts of glacier retreat in the tropical An-des, through the implemen-tation of specific pilot adaptation activities that illustrate the costs and ben-efits of adaptation

Climate trend analysis National-scale climate scenario projections based on Global Circulation Model from the Earth Simulator at MRI Japan; statistical and dynamic down-scaling

Local-scale analysis of socio-economic conditions

Mostly disciplinary scope

Problems framed under a pilot project approach, de-fined within a specific water catchment.

Connecting knowledge gen-eration community with large users (such as water utility companies, national to municipal governments, NGOs); and beneficiaries (farmers, small-scale irriga-tion and water supply asso-ciations)

Scientific information gen-erated by the project used to better inform specific pilot adaptation measures, as input for policy and planning, such as: provin-cial government’s climate change strategies, land-use planning, and catchment management plans. Scientific information also used for raising awareness through different mass media products

PACCb

2008–2012 Swiss Agency for

Development and Cooperation, national/provincial gov’ts

Peru: Cusco and Apurı´-mac regions

Disasters, water resources, agriculture

To understand vulnerabil-ities related to climate change, develop tools and capacities to address cli-mate impacts

Climate trend and climate sce-nario projections at regional scale

Sector/system analysis per scientific discipline

Interdisciplinary interaction re-latively late in process

Broad categories of systems defined, reconciliation pro-cess between science and policy during workshops, limited problem framing

Scientific information used for provincial government adaptation strategies. Practical adaptation actions at local level with limited connection to scientific stu-dies.

Glaciaresc

2011–2015 Swiss Agency for

Development and Cooperation, national/provincial gov’ts

Peru, Ancash and Cusco regions

Glaciers, water resources, disasters, agriculture

To improve the adaptation capacities in view of glacier retreat in Peru, including the institutional, technical and research capacities

Climate and glacier trend ana-lysis at regional level

Scientific basis of early warning and risk reduction at local level Interdisciplinary analysis of so-cio-environmental systems (incl. ethnographic studies)

Problem framing between science and executing NGO, with involvement of muni-cipalities and local commu-nities.

Due to data limitations sys-tems initially only broadly defined, later adjusted.

Local-scale climate risk studies used for adaptation measures; limited use of regional-scale scientific in-formation INAPd 2006–2011 World Bank, national gov’t Colombia Macizo de Chingaza, National Park Nevados Glaciers, water resources, agriculture, health

To support Colombia’s ef-forts to define and imple-m e n t s p e c i fi c p i l o t adaptation measures and policy options to meet the anticipated impacts of cli-mate change

Efforts focused on high mountain ecosystems and Caribean islands

Climate trend analysis National-scale climate scenario projections based on Global Circulation Model from the Earth Simulator at MRI Japan; statistical and dynamic down-scaling

Hydrological monitoring and modeling

Disciplinary and partly inter-disciplinary scope

System definition in line with those defined for Co-lombia’s National Commu-nication to the UNFCCC, i.e. high mountain ecosystems, island areas and human health.

Scientific committee invol-ving experts from different disciplines accompanying the process

Scientific information used for, inter alia: (i) mainte-nance of environmental services and adaptive land-use planning in high moun-tain ecosystems; (ii) pre-paration of malaria and dengue early warning sys-tem

Community-based project components with broad connection to scientific in-formation

Macizo Colombiano 2008–2011

United Nations (UNEP, UNICEF, FAO), national gov’t

Colombia, Upper Cauca river catchment Water resources, ecosystems

To support integrated management of ecosystems in view of climate adapta-tion, land-use planning and empowerment of local ac-tors

National level: generation of climate and water resrouce scenarios; analysis of vulner-abilities and risks in selected catchments

Local level: basic information for early warning, including the use of new hydrometerological stations

Framing and problem defi-nition in participatory ap-p r o a c h e s w i t h l o c a l communities

Knowledge dialog between different actors

Vulnerability and risk ana-lysis information and re-lated knowledge dialog used for implementation with local communities

PPCRe

Ongoing World Bank,

national gov’t

Bolivia, Rio Grande basin

Water resources

To support the implemen-tation of Bolivia’s Strategic Program for Climate Resili-ence by strengthening in-stitutional capacity; and supporting its implementa-tion river basin manage-m e n t i n t h r e e p i l o t catchments of the Rio Grande basin

Evaluation of water resources in selected river basins under climate change scenarios River basin plans formulated through a participatory process taking into account the impacts of climate change for water resources management for dif-ferent stakeholders

Framing and scoping by participatory process invol-ving a wide range of stake-holders of the river basin

Use of hydro-meteorologi-cal information (observa-tions and projec(observa-tions) for river basin planning pro-cess of stakeholders Prioritization of implemen-tation activities according to their relevance for cli-mate adaptation IMACCf 2012–2014 Interamerican Development Bank, Ministry of Environment of Peru Peru´, Ancash Water resources, disasters

To improve people’s lives, to implement adaptation measures with public sector i n v o l v e m e n t a n d t o strengthen regional capaci-ties

Climate and glacier trend ana-lysis at regional level

High mountain ecosystem ana-lysis, including water quality and biological indicators Social and institutional system analysis; scientific basis of early warning and risk reduc-tion at local level

Problem framing between science and public sector (local and national level) with involvement of local municipalities and commu-nities

Watershed-scale climate risk studies used for adap-tation measures

Remote sensing studies to support management plans Development of public in-vestment projects

a _{PRAA: Adaptation to the Impact of Rapid Glacier Retreat in the Tropical Andes.} b_{PACC: Climate Change Adaptation Program, Peru.}

c _{Glaciares project: Climate change adaptation and disaster risk reduction due to deglaciation of the Andes Cordilleras, Peru.} d_{INAP: Integrated National Adaptation Project, Colombia.}

e _{PPCR: Pilot Project on Climate Resilience, Bolivia.}

f_{IMACC: Implementation of adaptation measures in selected catchments, Peru.}

the Andes countries (for a complete documentation see (AndesPlus,2013).

Where possible the policy needs in terms of scientific

informationwereconsideredtohighlightthelinksbetween

scienceandpolicy.Duetolimitationsinspacewecanprovide

hereonlyasummarywhichshouldbecomprehensive,yetto

somedegreeneedstobeselective,consideringthebreadthof

researchinallthedifferenttopics.

4.1. Climatology

Inlinewiththegeneralconceptofthescientificassessment

wedistinguishbetweenclimateobservationsandprojections.

Thepastclimaterecordisafundamentalbasisforanyanalysis

inviewofclimatechangeadaptationandatthesametimea

challengeintheAndes,primarilybecauseofthe

aforemen-tioned limited availability of long-term, quality-checked

climatestationdata.Tocopewithlimiteddatabases(Schwarb

et al.,2011)presentedtools and methodsforchecking and

homogenizationofclimatestationdata asprovidedby the

national meteorological services, and related to ongoing

initiatives on climate services ofthe World Meteorological

Organization(WMO).

Additionaldatafromsatelliteproductssuchasfromthe

TropicalRainfallMeasurementMission(TRMM)(Lavadoetal.,

2009;Scheelet al., 2011;Zulkafliet al.,2013),or reanalysis

products(Hoferetal.,2010;Salzmannetal.,2013)havebeen

usedintheAndestocomplementthegroundstationdata.All

thesedatasetshavecertainpotentialandlimitationswhose

discussionisbeyondthescopeofthisstudybutdataquality

andreliabilityisacentralissuetoallofthem.

For climate projections both dynamical and statistical

downscaling of General Circulation Models has been

per-formedfortheAndesregion,forinstancewiththePRECISor

WRFmodels(Sanabriaetal.,2009;UrrutiaandVuille,2009),or

statisticalmethods(MinvielleandGarreaud,2011).

Neverthe-less, for statistical downscaling long-term, plausible and

homogeneoushistoricaldataseriesareanimperative

require-ment.However,inthetropicalAndeslongdecadaltimeseries

without significant gaps or inhomogeneities are quite rare

Fig.3–Exampleofamethodologicalsequenceofascientificassessment.Level1referstoobservationbasedstudieswhile

level2primarilyincludesprojection-basedapproaches.Themaingreenboxcorrespondsto,andexpandsonstage2

(middlebox)inFig.2.Tomaintainacomprehensiveperspectiveoftheprocesstheintersectionwiththecomponentswith

strongpolicyandstakeholderinvolvementareshownaswell(topandbottomblueboxes).(Forinterpretationofthe

referencestocolorinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.)

(Schwarbetal.,2011)andcancomplicateadequatestatistical downscaling.

Climateextremesareessential fortheunderstandingof

manyclimateimpacts, andthereforeimportanttoanalyze.

Standard extremeweatherand climateindiceswere

devel-opedbyClivar(ClimateVariabilityandPredictabilityProject),

and (Skansiet al., 2013) have found warmingand wetting

signalsoverSouth America,yet withnoclear precipitation

trendsoverthetropicalAndes.However,fromatotalof287

stations,forthetropicalAndesonly20stationsover2000masl

and12over3000maslwereavailableforanalysis,highlighting

theexacerbationofdatascarcityathigherelevations.

4.2. Waterresources

WaterresourcesareoffundamentalimportanceintheAndes.

Thereiswidespreadconcernthatimpactsofclimatechange

willnegativelyaffectwaterresourceavailability,andpeople

andeconomythatrely onthem(Buryet al.,2011;Buytaert

etal.,2011;Vergaraetal.,2007).Aparticularfocusisonareas

thatdependonmeltwaterfromglaciers,especiallyduringthe

dry season in large parts of Peru, Bolivia and Ecuador.

However,theexactcontributionofglacierandsnowmeltto

runoffisstillinsufficientlyunderstood.

Hence,bothforscienceandpolicyacriticalquestionishow

to determine present and future water resources and the

effectsof climatechangewith afocus on changingglacier

contribution to runoff. Glaciological studies in the Andes

investigated processes and changes across various scales,

fromdocumentingglacierchangesfromColombiatoBolivia

(Ceballosetal.,2006;Salzmannetal.,2013;SilverioandJaquet, 2005;Sorucoetal.,2009),toenergyandmassbalancestudies

onindividualglaciers(Favieretal.,2004;Francouetal.,1995;

Wagnon et al., 1999; Winkler et al., 2009). A number of

hydrologicalmodelshavebeenappliedtosimulatingrunoff

fromglacierfedcatchments(Baraeretal.,2012;Kaseretal.,

2003)includinganumberofstudiesthathaveusedcoupled

climatemodeldatatorunhydrological modelstoestimate

changes in future water resources (Andres et al., 2014;

Chevallieretal.,2011;Condometal.,2012;Juenetal.,2007; Lavadoetal.,2011).Amainlimitationforsuchstudiesisscarce

dataavailability,especiallyconcerningrainandrunoffgauges

overclimaticallyandhydrologicallysignificanttimeperiods.

Inaddition towaterquantity, thequality ofwaterisof

majorimportancetolocalpeople,andisinfactamajorissue,

e.g.inrelationwithminingactivities(BebbingtonandBury,

2009). Waterisalsoaneminentlysocialissueandconflicts

overwaterhavearisen inmanypartsoftheAndes. Power

discrepancies and inequalities play an important roleand

need to be addressedin water policies and regulations to

reduceandavoidconflicts(Careyetal.,2012a,b;Lynch,2012).

Fromapolicyperspectivewaterresourcemanagementisa

maininterest,recognizingthevariousdriversandstressors,

both natural and human, that act on water resources.

Available documents and experiences reported during the

workshops suggest that recent years have seen enhanced

effortsonwaterresourcemanagementbybothscienceand

policy,withanincreasingnumberofstudiesthatlinkacross

scientific disciplines and institutions (Carey et al., 2014;

Harriman,2014).

4.3. Agricultureandfoodsecurity

Agricultural production and food security under changing

climaticconditionsisofessential importanceinthe Andes,

fromsubsistencefarmingintheAltiplanoregionofPeruand

Boliviatocashcropssuchasquinoaorcoffeeinthecordillerasof

PeruandColombia.Policyisinterestedinchangingcropyields

under climate change, the importance of crop pests and

diseases,andmoregenerallyhowtoassesseconomicoutput

and food security. Climate variability and change act as

additional stressors on top of historically developed social,

political and economic structures, including land rights,

distribution,accessandmarginalization.Inmanyregions(e.g.

theAltiplano)agriculturalproductionisalreadyatitslimitsand

additionalstressorscantipthesystem(Sietzetal.,2012).

Scientificcontributionsinthisfieldincludetheapplication

ofcropmodelsconsidering phenological phasesand water

requirements(Geertsetal.,2006,2008;Stedutoetal.,2009),or

the assessment of vectors forpests and diseases (Gregory

etal.,2009).Pilotadaptationexperienceshavebeengenerated

at many locations,from Colombia through Peru to Bolivia

(Perezetal.,2011;PNUD,2011).

Large-scale integrated assessment models focusing on

agricultureproductionunderclimatechangescenarios,such

asapplied in severalother regions worldwide(Rosenzweig

etal.,2013)areyetlargelymissingintheAndes.

4.4. Naturalhazardsandrisks

Policy makers at different administrative levels and other

stakeholdersneedtoknowwhereandtowhatextentnatural

hazardsandrisksarelocated,bothinpresentandfuture,and

towhatextentclimatechangeexacerbatesthefrequencyor

magnitudeofclimaticextremeeventsand

hydrometeorologi-calhazardsingeneral.IntheAndes,sciencehasaddressed

theseissuesonlyinalimitedwaysofar.Disasterdatabases

areanimportanttooltobetterunderstandwhereandwhena

specificdisaster occurs.Across theAndes,Desinventar isa

primarydatabase(DesInventar,2013;Marulandaetal.,2010),

complemented by disaster databases at the nationallevel.

However,consistencyofthedisasterrecordacrosstimeand

differentdatasourcesislimitedandrepresentsabarriertoa

morecomprehensiveunderstanding(Huggeletal.,2014).

Awealth ofcasestudiesexistonsusceptibility, triggers,

andeffects ofalarge rangeofhazardprocesses, including

rainfall triggered landslides and debrisflows, ice/rock

ava-lanches,glacierlakeoutburstandrainfallfloods(e.g.Evans

etal.,2009;Piersonetal.,1990;Terlien,1997;Thouretetal., 1990;Vilimeketal.,2005).Studiesthatfocusonpolicyrelevant

aspectssuchashazardandriskmapping,orproduceabasis

forearlywarningsystemsaremorerarebutcorresponding

effortshaverecentlybeenstrengthened(Huggeletal.,2010;

Ku¨nzleretal.,2012;Schneideretal.,2014;Thouretetal.,2013).

Due to widespread poverty in the Andes, the social

dimensionofvulnerabilitytoclimatichazardsisofparticular

interestto policyand developmentplanners. Infact,there

existsalargeamountofstudiesonvulnerabilities,andsocial

andinstitutionalaspectsofriskperceptionandmanagement

intheAndesregion(e.g.CardonaandCarren˜o,2011;Nathan,

2008) but onlya fewhave been publishedin international

journals while the majorityarestudies performed bylocal

non-governmentalorganizationsoruniversitiesthatareoften

difficulttoaccess.

4.5. Ecosystems

Ecosystemservicescanbuildanimportantpillarinclimate

adaptation.Howthesecanbepreservedorenhancedunder

multiple stresses and drivers, including effects of climate

change,tosecuresupportingwell-being,reducingpovertyand

risks, and providing sustainable goods and services is

therefore of majorinterest (Carpenter et al., 2009; Fischlin

et al., 2007). Several critical Andean ecosystems, such as

paramos(highAndeanmoorlands)orAndeanforests,deliver

significant benefits in terms of water supply, water flow

regulation,anderosioncontrol,tonamebutafew(Anderson

et al.,2011; Buytaertetal., 2011). Agood understandingof

these services, and howthey willbe impacted by climate

changeandeconomicactivity,bothqualitativelyand

quanti-tatively,becomesthuscrucialintheadaptationdebate.

Overthe pastyears, severalprojectsinthe Andes have

generatedtangibleexperiencesinecosystemadaptationand

in strengthening the resilience of ecosystems and their

services. Someofthe mostoutstandingprojectshave been

theIntegratedNationalAdaptationPilot(INAP)inColombia,or

theaforementionedPRAAprojectinBolivia,EcuadorandPeru,

which have been based on science and local knowledge,

fosteringknowledgegenerationanddissemination(Andrade

etal.,2011)(Table1).Bothinitiativesappliedaparticipatory

and culturally sensitiveapproach topromoteresilient

eco-systemsandprovidebenefitstovulnerablepeople.Inviewofa

systematicevaluationofadaptationoptions(seealsoSection

5)modelswouldberequiredtointegratebio-physical,social

andeconomicaspectsinorder toprovide stakeholdersand

decision makerswithsufficient knowledge.While inother

regionsofthe worldintegratedassessment modelssuchas

GLOBIOM, G4M or EPIC have been used to assess future

scenarios of adaptation options (e.g. strict biodiversity

protection, nodeforestation) and tosupport corresponding

policydecisions(Havlı´ket al.,2011;Strassburgetal.,2012),

suchmethodsarebasicallystillmissingforthetropicalAndes.

Suchmodelscanalsodeliverbetterestimatesofecosystem

services to set in place appropriate offset schemes, for

example the Socio Bosque program in Ecuador, or the

environmental service payments in Bogota to protect the

nearbyparamoinChingaza.Theseprogramsaswellaspilot

experienceswiththeEPICmodelmadeinthecontextofthis

study indicate the importance of considering the specific

characteristicsoflocalAndeanecosystemsandagricultural

systemsfromsubsistencetoexport-orientedfarming,butalso

aspectsoflimitedavailabledatawhenapplyingsuchmodels.

4.6. Social,cultural,economicandpoliticalaspects

Thepreviousthematicaspectssuchaswaterresources,food

security or natural hazards and risks have a fundamental

socialcomponent.Forinstance,howtorespondtowhichrisks

doesnotonlydependontherisksthemselvesbutiscritically

shapedbythedynamiccultural,social,economic,politicaland

environmentalcontext.Thesocialsciencestudiestherefore

needtorunacrossthethematicaspects,andassuchcanexert

acertainintegrativerole.Withinthisprocess,theimpactsof

globalandnational-scaledriversonlocalprocessesandthe

roleoflocalandscientificknowledgeforclimateadaptation

needto be reasonably understood. Thenumber ofstudies

concerningthesocialandculturaldimensionisincreasingbut

yetthereisaneedtofurtherdeepentheunderstandingofthese

topics, particularly in their specific local context where

concreteadaptationstrategiesareputintopractice.

There is a considerable interest of policy makers and

adaptationprojectsforsocalled‘toolkits’toassessperceptions

andvulnerabilitiesincommunitybasedapproaches,suchas

e.g.CRISTALandCVCA(Daze´ etal.,2009;IISD,2012).Thesetools

offer a rapid appraisal to analyze risks and vulnerabilities

withincommunitiesinaspecificcontext.Nevertheless,data

collectionmethodsofthesetoolkitsareoftennotsuitablefor

adequatelyunderstandingdecision-makingprocesses,power

relationsand openand/orlatentsocial or politicalconflicts

withinorbetweeninstitutions.Suchissuesareoftenrelatedto

trust and distrust among the actors and institutions, and

specific interests and cultural values, within or between

institutions,andtheirunderstandingthusrequires

consider-ationofthefullrangeofactors,includingthepowerlessones.

Forprovidingasoundscientificbasisforclimateadaptation

though,thereisaneedtoapplymorein-depthmethodssuch

as ethnographies, participant observation, narrative and

historicalanalysis,or mentalmodelstoexploreknowledge

systems(Adgeretal.,2013;Barnesetal.,2013).Thesemethods

allow to identify social, cultural, political and economic

aspectsthatarecrucialfortheadaptationprocess,butalso

particularlyimportant forjoint knowledge production

pro-cessesbecausetheyfacilitatetherecognitionofdifferencesin

actorperspectives(HeggerandDieperink,2014).Furthermore,

amutualunderstandingofthecultural,social,politicaland

economicbackground,oftheconstraintsthattheactorsface

onthe adaptationpathwayand theperceptionsthey have,

enhancestrustamongthem. However,asstressedbyAdger

et al., 2013 there is no simple blueprint in terms of

methodology,andtheconsiderationofmultipleandmarginal

voicesintodecision-makingishighlychallenging.

5.

Adaptation

options

and

implementation

Developmentandevaluationofadaptationoptions,andtheir

implementationrepresentsthethirdstageoftheframework

(Fig.2).Itisabroad,complexandyoungfieldbutwitharapidly

growingbodyofliteraturewhichwecannotcapturehereinits

full breadth. Obviously, the implementation of adaptation

measuresisbeyondthescientificcontributiontoadaptation

andthereforebasicallyalsobeyondthescopeofthisstudy.

Herewefocusinsteadontheanalysisofrecentmajorclimate

adaptationprojectsinthetropicalAndeswheretherehasbeen

asciencecontribution.Weconsiderthisparticularly

worth-while because the documentation of this process across

adaptationprojectsinthisregionisunderdevelopeddespite

recenteffortsintheIPCC(Magrinetal.,2014).

AsindicatedinFig.2adaptationdecisionsarenottakenina

vacuumbutinadynamicsocial,cultural,economic,political

andlegalcontext. Climate(change)typically representsone

amongmanydriversofchange,andstressors,resultinginrisks

(oropportunities)that are perceiveddifferentlybydifferent

actorsandstakeholders.Mostprojectsanalyzedhere

concen-trated on adaptation measures that can be viewed as

vulnerabilityreduction,andenhancementofadaptive

capaci-tiesandgovernance.Inseveralcasesclimatevariabilityand

extremeswereconsideredamoreimmediatepriorityforthe

stakeholders as compared to longer term climate change

projections. All the projects have local and national scale

components and address multiple sectors, with water

resourcesbeingadominantissueinmanyprojects,reflecting

theimportanceofwaterfortheAndesregion.Inseveralprojects

scientificinformationwasusedtosupportthedevelopmentof

nationalorprovincialadaptationstrategies.Effortswerealso

madetousescientificinformationforevaluationofadaptation

optionsbutdifficultieswerefound atthelocallevel,mainly

because scientific assessments struggled with the lack of

availabledata,especiallyoverlongerperiodsoftime.Instead,

severalprojectsappliedmorepracticalapproaches,applying

participativemethodsandtoolswithlocalcommunities(see

alsoSection4.6).Especiallyforwaterresources,andagriculture

andfoodsecurity,manyprojectsimplementedsuchadaptation

strategies with local communities but with limited or no

connectiontoscience.Yet,therearesome exampleswhere

sciencehassubstantiallycontributedtotheimplementationof

adaptationmeasures,e.g.forearlywarningandriskreduction

measures(Schneideretal.,2014).Furthermore,manyprojects

implementedadaptationmeasurestoimproveenvironmental

monitoringnetworks.Scienceanddifferentlevelsofpolicyand

otheractorsusuallyjointlyevaluatedoptions,anddecidedon,

whereandwhatsortofequipmenttoinstall.

Oneimportantbarriertosuccessfuladaptation

implemen-tationwehaveseenacrossmostprojectsaredifferenttime

framesbetweenpolicyandotheractors,andscience.Policy

anddecisionmakersatvariousadministrativelevelsareoften

insufficiently awareof thetime neededto producea solid

scientificbasis,andhence,scientistswereunabletomeetthe

time frames expected from policy makers or adaptation

practitioners. Thetimeperiods inFig. 2areindicativeand

essentiallydependontherespectivecontextandprojects.In

the Andescountries, more than inEurope or the U.S., the

termsoflocal,provincialandnationalgovernmentsexerta

majorinfluenceonadaptationplanningandimplementation,

and science cannot act independentlyof those terms in a

concerted adaptation process. The terms are typically 4–5

years,andideallyareconsideredintheprocess.Eventually,it

is important to recognize the iterative character of the

adaptation option and implementation process. This is

especiallyimportantinviewoftheuncertaintiesinherentto

the scientific knowledge and socio-economic processes

(Websterand Jian,2011),and thus representsa strategyto

manageuncertainties(Gersoniusetal.,2013;Hallegatte,2009)

6.

Discussion

6.1. ReflectiononAndesexperiences

The framework for the science contribution in climate

adaptationthatwasdevelopedwiththeAndeangovernments

distinguishes three main stages (Fig. 2): (1) framing and

problem definition; (2) scientific assessments of climate

impacts, vulnerabilities and risks with an interdisciplinary

perspective;and(3)adaptationoptionevaluationand

imple-mentation.Thisframeworkisbasicallyinlinewithsimilar

conceptsthathavebeenassessedanddevelopedrecentlyby

theIPCC(IPCC,2014),wherethecompleteadaptationprocess

isframed underarisk managementscheme withmultiple

feedbacksanditerations.Duringtheassessmentofthe

state-of-the-artinthisfieldinthetropicalAndesregionwehave

identifiedthefollowinggapsandkeyissues,asrelatedtothe

aforementionedstagesoftheframework:

Thereisaneedformoreconsistencyincommonproblem

framing,terminology,objectives,timeframes,andtypeof

resultsbetweenscience,andpolicyandlocalactors(stage1);

There is a need for a more sustained interaction and

communicationofsciencewiththediverserangeofactors

thathaveastakeinadaptation(stages1and3);

Limitationsindataavailabilityanddatasharingcontinueto

persist,bothatcountryandregionalAndeanlevel.

Equip-mentforenvironmentalmonitoringisofteninsufficientbut

recentlyimprovingduetomultipleefforts(stage2);

The interdisciplinary scope of science studies and the

interaction among disciplines arepredominantly notyet

developedtoalevelthatallowsscientistsandpolicymakers

toapproachtheproblemsintheirfullcomplexity(stage2);

There isalackofappropriate monitoringandevaluation

tools,whichrendersthelearningandstock-takingprocess

ofthedifferentexperienceschallenging(stage3).

SomeoftheabovepointsarerelevantbeyondtheAndes

regionandaretakenoninthenextsection.Otherissuessuch

aslimitedenvironmentalmonitoringdataandnetworksare

alsoalimitationinmanyotherregions,andexacerbatedin

mountainandhighmountainregions(Salzmannetal.,2014).

Methodsthatweredevelopedinregionswithdense

monitor-ingnetworksneedtobeadapted.Examplesofstudieswith

methods adapted to the Andean context in the field of

climatologyandwaterresources,forinstance,werepresented

by(Buytaertetal.,2009;Lavadoetal.,2009;Salzmannetal., 2013;Scheeletal.,2011)intermsofcombinationofmultiple

datasets,andby(BuytaertandBeven,2011;Chevallieretal.,

2011;Condometal.,2012)intermsofappropriatecomplexity

formodelingstudies.

The identification of methods that are adapted to the

Andean context, or locally developed, was one of the

important issues for the scientific assessment of impacts,

vulnerabilities and risks (stage 2, Fig. 3). The process of

achieving a consensus between Andes and Europe based

expertsfromscienceandgovernmentsonmethodologiesand

conceptsineachthematicfieldwasnotstraightforwardanda

challengingscience-policyprocessinitself.Itwasstartedbya

presentation of an initial compilation of methods by the

internationalresearchteamtotheAndesexperts,primarily

along disciplinary lines; followed by a comparison with

locally/regionally(Andes)availablemethods.Methods

adjust-edto the Andean context werethen jointly identifiedand

definedthroughareviewprocessofseveraldraftdocuments,

which concluded in a final document that was eventually

officially approved by the four national governments of

Colombia,Ecuador,PeruandBolivia(AndesPlus,2013).

Theappropriateconsiderationofarichandsteadilygrowing

Andeanexperiencerepresentedamajorchallengethroughall

stages(1)–(3).However,muchoftheknowledgeproductionand

adaptation processesin the Andesare, however,not

suffi-ciently documented and rarely published in international

journals.Well-established,long-term,sustainablemonitoring

andevaluationschemesofexperiencesarealsobroadlylacking

asindicated above.Therecapitulationoftheseexperiences

with local experts was a fundamental aspect but also a

challenge.Forinstance,manyadaptationexperienceshavea

practical focus, often implemented by non-governmental

organizations,withmethodsthatarenotnecessarilyinline

with standards ofscientific rigor.We have found that the

distinction between scientific and practical approaches in

adaptation is often not clearly defined, and hence, the

integrationofsuchdiverseapproachesisdemanding.

6.2. GeneralimplicationsofAndesexperiences

Adaptationplanningisoftencategorizedintotop-down and

bottom-upapproacheswherethefirstonetypicallyrefersto

climatescenarioandimpactdrivenassessments,andthelatter

startswithvulnerabilityassessments,oftenwithqualitative

andparticipatorymethods,basedonwhichadaptationoption

arethenevaluated(WilbyandDessai,2010).Recentresearch

suggeststhatinfacttheintegrationofthetwoapproachesisa

moresuccessfulmodel(Bhaveetal.,2014).Ourframeworkin

in-linewiththeseproposalsasitcombinesimportantelements

frombothapproaches.Someofthethematicscientific

assess-ments (see Figs. 2 and 3) include elementsof a top-down

approachwhiletheframingandproblem/objectivedefinitionis

reflectingaspectsofabottom-upapproach.Furthermore,the

integrativeandcross-thematicnatureofthesocialaspectsin

ourframeworkcontributetoensureanappropriate

consider-ation of the human dimension of adaptation. Top-down

approacheshavebeencriticizedtoneglectthisdimension.It

isinterestingtonotethatouranalysisofadaptationprojectsin

thetropicalAndes(Section5)hasshownthatoftensomesortof

blendedapproachisapplied,motivatedorconstrainedbythe

lack of local data. Some of those project activities were

decoupled from science. While actually not all types of

adaptationmayneedascientificcontributionweconsiderthat

science can oftenenhance the joint knowledge production

process.In-depthsocialsciencestudies,forinstance,areableto

shedlightonpowerrelations,interestsandvaluesofvarious

actors and institutions, thus contributing to socially more

concertedadaptation.However,wealsorecognizethatinthis

contextthescience-policyprocesscanbeparticularlydelicate

asimportantpoliticalinterestsmaybeatstake.

Moregenerally,wenotethatthereisarichandgrowing

literatureontheproductionofknowledgebasedonscientific

understandinganda rangeofothersourcesinan

environ-mentaland climateadaptation context.Scholarsreflect on

both theoretical and practical aspects, with a majority of

studiesbasedonexperiencesfromEuropeandNorthAmerica

(Cashetal.,2006;DillingandLemos,2011;Heggeretal.,2012; Lemos and Morehouse, 2005). Similar experiences from

regionssuchasSouth Americaareavailableyettoalesser

degreealthough theyare ofgreat importanceforregional,

nationalandlocaladaptationpolicy(Lynch,2012).

Recent research has identified a number of functional,

structural,andsocialbarrierstoajointproblemframingand

knowledgeproduction,thataccordinglymaybegroupedin(1)

divergentobjectives,needs,scope,andpriorities;(2)different

institutionalsettingsandstandards,andtimeframes;and(3)

differingculturalvalues,understanding,andmistrust(Hegger

etal.,2012;WeichselgartnerandKasperson,2010).

Concerningthefirstgroupofbarriersouranalysisofmajor

adaptationprojectsintheAndeshasshownthatinpractice

reconciliationofobjectivesandscopeisonlyrarelydonetothe

extentitwouldbebeneficial(Table1).Acomplicatingfactor

therebyisthattheaforementionedthreegroupsofinhibiting

factorsarehighlyinterdependent,i.e.reconcilingobjectives

andneedscannotbeachievedwithoutaddressingdifferences

suchasintermsofinstitutionalsettings,timeframes,cultural

valuesandtrust.

Whilesomeoftheseissuescanbeapproachedinamore

systematicway(e.g.timeframes),thisismuchmoredifficultfor

others(e.g.culturalvaluesandtrust)whichmakestheprocess

notfullycontrollableorpredictable.IntheAndes,aswellasin

otherregions,climateadaptationtakesplaceontheregional,

national or local level, with a variety of actors (e.g. CAN,

internationalcooperation,developmentbanksontheregional

level;localgovernments/communities,non-governmental

or-ganizationsonthe local level).Accordingly, thereisawide

rangeofaspectstoconsiderwhenobjectivesandneedsare

reconciledonthedifferentlevelsandbetweenactors.Inour

framework(Figs.2and3)wehavenotexplicitlydistinguished

betweenthedifferentadministrativeandspatiallevels,mainly

tomaintainitsgeneralapplicabilityand constrainthe

com-plexity.Infact,giventhecurrentdiscussionsinliterature,and

basedonourownexperienceinotherregions,weanticipate

thattheframeworkmaybeusefulbeyondtheAndesregion.

7.

Conclusions

and

recommendations

Inthispaper,wehavepresentedandanalyzedaframework

forthesciencecontributionembeddedinthebroaderclimate

adaptation process. It was developed in a joint process

involving representatives of the four governments of

Colombia, Ecuador, Peru and Bolivia, the CAN, the World

Bank,andscientistsbothfromtheAndesregionandEurope.

Three main stages, from problem framing and scientific

assessments to adaptation implementation, are

distin-guished.Appropriatelyputtingtheprocessintotherespective

dynamicsocial,cultural,economic,politicalandlegalcontext

isoffundamentalimportancebuthighlychallenging.Clearly,

moreresearchonanalyzinganddocumentingsuchprocesses

andexperiencesisneeded.

Theconsistencyofmanyelementsanalyzedhereforthe

AndeswiththerecentassessmentbyIPCC,2014indicatesthat

whileadaptionisoftenalocalprocess,aspectsofscience-policy

interactionintheframeofadaptationareofmoreuniversal

significance.Specifically,ourstudyunderlinestheimportance

ofjoint problemframing, definition ofsocio-environmental

systems,timeframes,andamoreintenseinteractionofsocial

andphysicalclimateand impactsciences.Scientifically,the

scarcityofenvironmental,socialandeconomicdatainregions

liketheAndescontinuetorepresentalimitationtoadaptation.

Investmentsintosocio-environmentalmonitoringcan

there-forebeseenasaviablelow-regretadaptationmeasures.

Acknowledgements

ThisstudywasundertakeninthecontextoftheAndesPlus

initiative of the PRAA project, coordinated by the Andean

CommunityofNationsandtheWorldBank,withfundsfrom

theGlobalEnvironmentFacility.TheSwissFederalOfficefor

theEnvironmentandtheSwissAgencyforDevelopmentand

Cooperationprovidedsubstantialsupportat variousstages.

Weappreciatethecollaborationandinteractionwithalarge

numberofexpertsfromtechnicalandscientificinstitutionsof

thenationalgovernmentsofBoliviaColombia,Ecuadorand

Peru,andnon-governmentalorganizations.

r

e

f

e

r

e

n

c

e

s

Adger,W.N.,Barnett,J.,Brown,K.,Marshall,N.,O’Brien,K.,

2013.Culturaldimensionsofclimatechangeimpactsand

adaptation.Nat.Clim.Change3,112–117,http://dx.doi.org/

10.1038/nclimate1666.

Anderson,E.P.,Marengo,J.,Villalba,R.,Halloy,S.,Young,B.,

Cordero,D.,Gast,F.,Jaimes,E.,Ruiz,D.,2011.Consequences

ofclimatechangeforecosystemsandecosystemservicesin theTropicalAndes.In:ClimateChangeandBiodiversityin theTropicalAndes.MacArthurFoundation,Inter-American InstituteforGlobalChangeResearch(IAI),Scientific CommitteeonProblemsoftheEnvironment(SCOPE),1–5.

AndesPlus,2013.Metodologı´asparalaformulacio´ndeLı´neasde

BaseyMedidasdeAdaptacio´nalCambioClima´ticoen EcosistemasdeAltaMontan˜a.UniversityofZurich,Andean CommunityofNations,Zurich/Lima.

Andrade,A.,Co´rdoba,R.,Dave,R.,Girot,P.,Herrera-F,B.,Munroe,

R.,Oglethorpe,J.,Pramova,E.,Watson,J.,Vergara,W.,2011.

DraftPrinciplesandGuidelinesforIntegrating Ecosystem-BasedApproachestoAdaptationinProjectandPolicyDesign: ADiscussionDocument.TropicalAgriculturalResearchand HigherEducationCenter(CATIE),InternationalUnionforthe ConservationofNature(IUCN),Turrialba,CostaRica.

Andres,N.,VegasGaldos,F.,LavadoCasimiro,W.S.,Zappa,M.,

2014.Waterresourcesandclimatechangeimpactmodelling

onadailytimescaleinthePeruvianAndes.Hydrol.Sci.J.0,

http://dx.doi.org/10.1080/02626667.2013.862336(inpress).

Baraer,M.,Mark,B.G.,McKenzie,J.M.,Condom,T.,Bury,J.,Huh,

K.-I.,Portocarrero,C.,Go´mez,J.,Rathay,S.,2012.Glacier

recessionandwaterresourcesinPeru’sCordilleraBlanca.J. Glaciol.58,134–150.

Barnes,J.,Dove,M.,Lahsen,M.,Mathews,A.,McElwee,P.,

McIntosh,R.,Moore,F.,O’Reilly,J.,Orlove,B.,Puri,R.,Weiss,

H.,Yager,K.,2013.Contributionofanthropologytothestudy

ofclimatechange.Nat.Clim.Change3,541–544,http://

dx.doi.org/10.1038/nclimate1775.

Bebbington,A.J.,Bury,J.T.,2009.Institutionalchallengesfor

miningandsustainabilityinPeru.Proc.Natl.Acad.Sci.106, 17296.

Bhave,A.G.,Mishra,A.,Raghuwanshi,N.S.,2014.Acombined

bottom-upandtop-downapproachforassessmentof

climatechangeadaptationoptions.J.Hydrol.518,150–161,

http://dx.doi.org/10.1016/j.jhydrol.2013.08.039.

Bielak,A.T.,Campbell,A.,Pope,S.,Schaefer,K.,Shaxson,L.,

2008.Fromsciencecommunicationtoknowledge

brokering:theshiftfromsciencepushtopolicypull.In: Cheng,D.,Claessens,M.,Gascoigne,T.,Metcalfe,J.,Schiele, B.,Shi,S.(Eds.),CommunicatingScienceinSocialContexts. Springer,TheNetherlands, pp.201–226.

Biesbroek,G.R.,Swart,R.J.,Carter,T.R.,Cowan,C.,Henrichs,T.,

Mela,H.,Morecroft,M.D.,Rey,D.,2010.Europeadaptsto

climatechange:comparingnationaladaptationstrategies.

GlobalEnviron.Change20,440–450,http://dx.doi.org/

10.1016/j.gloenvcha.2010.03.005.

Botzen,W.J.W.,VanDenBergh,J.,2008.Insuranceagainst

climatechangeandfloodingintheNetherlands:present, future,andcomparisonwithothercountries.RiskAnal.28, 413–426.

Brown,C.,Wilby,R.L.,2012.Analternateapproachtoassessing

climaterisks.EOSTrans.Am.Geophys.Union93,401.

Bury,J.T.,Mark,B.G.,McKenzie,J.M.,French,A.,Baraer,M.,Huh,

K.I.,ZapataLuyo,M.A.,Go´mezLo´pez,R.J.,2011.Glacier

recessionandhumanvulnerabilityintheYanamarey watershedofthecordilleraBlanca,Peru.Clim.Change105, 179–206.

Buytaert,W.,Beven,K.,2011.Modelsasmultipleworking

hypotheses:hydrologicalsimulationoftropicalalpine

wetlands.Hydrol.Process.25,1784–1799,http://dx.doi.org/

10.1002/hyp.7936.

Buytaert,W.,Ce´lleri,R.,Timbe,L.,2009.Predictingclimate

changeimpactsonwaterresourcesinthetropicalAndes: effectsofGCMuncertainty.Geophys.Res.Lett.36,L07406.

Buytaert,W.,Cuesta-Camacho,F.,Tobo´n,C.,Buytaert,W.,

Cuesta-Camacho,F.,Tobo´n,C.,2011.Potentialimpactsof

climatechangeontheenvironmentalservicesofhumid

tropicalalpineregions,Potentialimpactsofclimatechange

ontheenvironmentalservicesofhumidtropicalalpine

regions.GlobalEcol.Biogeogr.20,19–33,http://dx.doi.org/

10.1111/j.1466-8238.2010.00585.x.

CAN,2008.Elcambioclima´ticonotienefronteras.Impactodel

cambioclima´ticoenlaComunidadAndina.Secretaria GeneraldelaComunidadAndina..

Cardona,O.D.,Carren˜o,M.L.,2011.Updatingtheindicatorsfor

disasterriskandriskmanagementfortheAmericas.IDRiMJ.

1,http://dx.doi.org/10.5595/idrim.2011.0014.

Carey,M.,Baraer,M.,Mark,B.G.,French,A.,Bury,J.,Young,K.R.,

McKenzie,J.M.,2014.Towardhydro-socialmodeling:

merginghumanvariablesandthesocialscienceswith

climate-glacierrunoffmodels(SantaRiver,Peru).J.Hydrol.

518(PartA) 60–70,http://dx.doi.org/10.1016/

j.jhydrol.2013.11.006.

Carey,M.,French,A.,O’Brien,E.,2012a.Unintendedeffectsof

technologyonclimatechangeadaptation:anhistorical

analysisofwaterconflictsbelowAndeanGlaciers.J.Hist.

Geogr.38,181–191,http://dx.doi.org/10.1016/

j.jhg.2011.12.002.

Carey,M.,Huggel,C.,Bury,J.,Portocarrero,C.,Haeberli,W.,

2012b.Anintegratedsocio-environmentalframeworkfor

climatechangeadaptationandglacierhazardmanagement: lessonsfromLake513,CordilleraBlanca,Peru.Clim.Change 112,733–767.

Carpenter,S.R.,Mooney,H.A.,Agard,J.,Capistrano,D.,DeFries,

R.S.,Dı´az,S.,Dietz,T.,Duraiappah,A.K.,Oteng-Yeboah,A.,

Pereira,H.M.,Perrings,C.,Reid,W.V.,Sarukhan,J.,Scholes,

R.J.,Whyte,A.,2009.Scienceformanagingecosystem

services:beyondthemillenniumecosystemassessment.

ProcNatlAcadSci106,1305–1312,http://dx.doi.org/10.1073/

pnas.0808772106.

Cash,D.W.,Borck,J.C.,Patt,A.G.,2006.Counteringthe

loading-dockapproachtolinkingscienceanddecisionmaking

comparativeanalysisofElNin˜o/southernoscillation(ENSO)

forecastingsystems.Sci.Technol.HumanValues31,465– 494,http://dx.doi.org/10.1177/0162243906287547.

Cash,D.W.,Clark,W.C.,Alcock,F.,Dickson,N.M.,Eckley,N.,

Guston,D.H.,Ja¨ger,J.,Mitchell,R.B.,2003.Knowledge

systemsforsustainabledevelopment.Proc.Natl.Acad.Sci.

100,8086–8091,http://dx.doi.org/10.1073/pnas.1231332100.

Ceballos,J.L.,Euscategui,C.,Ramirez,J.,Canon,M.,Huggel,C.,

Haeberli,W.,Machguth,H.,2006.Fastshrinkageoftropical

glaciersinColombia.Ann.Glaciol.43,194–201.

Chevallier,P.,Pouyaud,B.,Suarez,W.,Condom,T.,2011.

Climatechangethreatstoenvironmentinthetropical

Andes:glaciersandwaterresources.Reg.Environ.Change

11,179–187,http://dx.doi.org/10.1007/s10113-010-0177-6.

Condom,T.,Escobar,M.,Purkey,D.,Pouget,J.C.,Suarez,W.,

Ramos,C.,Apaestegui,J.,Tacsi,A.,Gomez,J.,2012.

Simulatingtheimplicationsofglaciers’retreatforwater

management:acasestudyintheRioSantabasin,Peru.

WaterInt.37,442–459,http://dx.doi.org/10.1080/

02508060.2012.706773.

Daze´,A.,Ambrose,K.,Erhart,C.,2009.ClimateVulnerability

andCapacityAnalysisHandbook.CAREInternational.

DesInventar,2013.DesInventar–InventorySystemofthe

EffectsofDisasters.Corporacio´nOSSA,Cali,Colombia.

Dewulf,A.,2013.Contrastingframesinpolicydebateson

climatechangeadaptation.WileyInterdisciplinaryReviews.

Clim.Change4,321–330,http://dx.doi.org/10.1002/wcc.227.

Dilling,L.,Lemos,M.C.,2011.Creatingusablescience:

opportunitiesandconstraintsforclimateknowledgeuse

andtheirimplicationsforsciencepolicy.GlobalEnviron.

Change21,680–689,http://dx.doi.org/10.1016/

j.gloenvcha.2010.11.006.

Evans,S.G.,Bishop,N.F.,Smoll,L.F.,Murillo,P.V.,Delaney,K.B.,

Oliver-Smith,A.,2009.Are-examinationofthemechanism

andhumanimpactofcatastrophicmassflowsoriginatingon

NevadoHuascara´n,CordilleraBlanca,Peruin1962and1970.

Eng.Geol.108,96–118,http://dx.doi.org/10.1016/

j.enggeo.2009.06.020.

Favier,V.,Wagnon,P.,Ribstein,P.,2004.Glaciersoftheouter

andinnertropics:adifferentbehaviourbutacommon responsetoclimaticforcing.Geophys.Res.Lett.31,16.

Fischlin,A.,Midgley,G.,Price,J.,Leemans,R.,Gopal,B.,

Turley,C.,Rounsevell,M.,Dube,O.,Tarazona,J.,Velichko,

A.,2007.Ecosystems,theirproperties,goods,andservices.

In:Parry,M.L., Canziani,O.F., Palutikof,J.P., vander Linden,P.J., Hanson,C.E.(Eds.), ClimateChange2007: Impacts,AdaptationandVulnerability.Contributionof WorkingGroupIItotheFourthAssessmentReportofthe IntergovernmentalPanelonClimateChange. pp.211–272.

Ford,J.D.,Vanderbilt,W.,Berrang-Ford,L.,2012.Authorshipin

IPCCAR5anditsimplicationsforcontent:climatechange

andindigenouspopulationsinWGII.Clim.Change113,201–

213,http://dx.doi.org/10.1007/s10584-011-0350-z.

Francou,B.,Ribstein,P.,Wagnon,P.,Pouyaud,B.,1995.Annual

cycleofenergybalanceofZongoglacier,CordilleraReal, Bolivia.J.Geophys.Res.104,3907–3923.

Geerts,S.,Raes,D.,Garcia,M.,Condori,O.,Mamani,J.,Miranda,

R.,Cusicanqui,J.,Taboada,C.,Yucra,E.,Vacher,J.,2008.

Coulddeficitirrigationbeasustainablepracticeforquinoa (ChenopodiumquinoaWilld.)intheSouthernBolivian Altiplano? Agric.WaterManag.95,909–917.

Geerts,S.,Raes,D.,Garcia,M.,DelCastillo,C.,Buytaert,W.,

2006.Agro-climaticsuitabilitymappingforcropproduction

intheBolivianAltiplano:acasestudyforquinoa.Agric.For. Meteorol.139,399–412.

Gentle,P.,Maraseni,T.N.,2012.Climatechange,povertyand

livelihoods:adaptationpracticesbyruralmountain

communitiesinNepal.Environ.Sci.Policy21,24–34,http://

dx.doi.org/10.1016/j.envsci.2012.03.007.

Gersonius,B.,Ashley,R.,Pathirana,A.,Zevenbergen,C.,2013.

Climatechangeuncertainty:buildingflexibilityintowater

andfloodriskinfrastructure.Clim.Change116,411–423,

http://dx.doi.org/10.1007/s10584-012-0494-5.

Gregory,P.J.,Johnson,S.N.,Newton,A.C.,Ingram,J.S.I.,2009.

Integratingpestsandpathogensintotheclimatechange/

foodsecuritydebate.J.Exp.Bot.60,2827–2838,http://

dx.doi.org/10.1093/jxb/erp080.

Hallegatte,S.,2009.Strategiestoadapttoanuncertainclimate

change.GlobalEnviron.Change19,240–247.

Harriman,L.,2014.Wherewillthewatergo?Impactsof

acceleratedglaciermeltintheTropicalAndes:Article

reproducedfromUnitedNationsEnvironmentProgramme

(UNEP)GlobalEnvironmentalAlertService(GEAS). Environ.

Dev.,http://dx.doi.org/10.1016/j.envdev.2013.10.001(in press).

Havlı´k,P.,Schneider,U.A.,Schmid,E.,Bo¨ttcher,H.,Fritz,S.,

Skalsky´,R.,Aoki,K.,Cara,S.D.,Kindermann,G.,Kraxner,F.,

Leduc,S.,McCallum,I.,Mosnier,A.,Sauer,T.,Obersteiner,

M.,2011.Globalland-useimplicationsoffirstandsecond

generationbiofueltargets.EnergyPolicy39,5690–5702,

http://dx.doi.org/10.1016/j.enpol.2010.03.030.

Hegger,D.,Dieperink,C.,2014.Towardsuccessfuljoint

knowledgeproductionforclimatechangeadaptation:

lessonsfromsixregionalprojectsintheNetherlands.Ecol.

Soc.19,34,http://dx.doi.org/10.5751/ES-06453-190234.

Hegger,D.,Lamers,M.,VanZeijl-Rozema,A.,Dieperink,C.,

2012.Conceptualisingjointknowledgeproductionin

regionalclimatechangeadaptationprojects:success

conditionsandleversforaction.Environ.Sci.Policy18,52–

65,http://dx.doi.org/10.1016/j.envsci.2012.01.002.

Hofer,M.,Mo¨lg,T.,Marzeion,B.,Kaser,G.,2010.

Empirical-statisticaldownscalingofreanalysisdatatohigh-resolution

airtemperatureandspecifichumidityaboveaglacier

surface(CordilleraBlanca,Peru).J.Geophys.Res.:Atmos.115

,http://dx.doi.org/10.1029/2009JD012556(n/a–n/a).

Huggel,C.,Khabarov,N.,Obersteiner,M.,Ramı´rez,J.M.,2010.

Implementationandintegratednumericalmodelingofa

landslideearlywarningsystem:apilotstudyinColombia.

Nat.Hazards52,501–518,

http://dx.doi.org/10.1007/s11069-009-9393-0.

Huggel,C.,Raissig,A.,Rohrer,M.,Romero,G.,Diaz,A.,

Salzmann,N.,2014.Howusefulandreliablearedisaster

databasesinthecontextofclimateandglobalchange?A

comparativecasestudyanalysisinPeru.Nat.HazardsEarth

Syst.Sci.Discuss.2,4331–4362,http://dx.doi.org/10.5194/

nhessd-2-4331-2014.

Huggel,C.,Rohrer,M.,Calanca,P.,Salzmann,N.,Vergara,W.,

Quispe,N.,Ceballos,J.L.,2012.Earlywarningsystems:the

lastmileofadaptation.EosTrans.AGU93,209,http://

dx.doi.org/10.1029/2012EO220001.

IISD,2012.CommunityBasedRiskScreeningTool–Adaptation

andLivelihoods.CRISTALUserManualVersion5.The InternationalInstituteforSustainableDevelopment, Winnipeg,MB,Canada.

IPCC,2014.SummaryforPolicymakers.WorkingGroupII

ContributiontotheIPCCFifthAssessmentReportClimate Change2014:Impacts,AdaptationandVulnerability. CambridgeUniversityPress,Cambridge,UK.

Juen,I.,Kaser,G.,Georges,C.,2007.Modellingobservedand

futurerunofffromaglacierizedtropicalcatchment

(CordilleraBlanca,Peru).GlobalPlanet.Change59,

37–48,http://dx.doi.org/10.1016/j.gloplacha.2006.11.038.

Kaser,G.,Juen,I.,Georges,C.,Go´mez,J.,Tamayo,W.,2003.The

impactofglaciersontherunoffandthereconstructionof massbalancehistoryfromhydrologicaldatainthetropical CordilleraBlanca,Peru.J.Hydrol.282,130–144.