Rights / License: Research Collection In Copyright - …1995/eth... · Prof. Dr. Roland W. Scholz,...

Research Collection

Doctoral Thesis

Decision-processes concerning the management of ecosystemservices for ecosystem-based adaptation to climate change

Author(s): Vignola, Raffaele

Publication Date: 2010

Permanent Link: https://doi.org/10.3929/ethz-a-006207064

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

https://doi.org/10.3929/ethz-a-006207064

http://rightsstatements.org/page/InC-NC/1.0/

https://www.research-collection.ethz.ch

https://www.research-collection.ethz.ch/terms-of-use

DISS.ETHNO.19154

DECISION‐PROCESSESCONCERNINGTHEMANAGEMENTOFECOSYSTEM

SERVICESFORECOSYSTEM‐BASEDADAPTATIONTOCLIMATECHANGE

Adissertationsubmittedtothe

ETHZURICH

forthedegreeof

DoctorofSciences

presentedby

RAFFAELEVIGNOLA

Agr.Eng.,UniversityofFlorence,Italy

bornFebruary4th,1969

CitizenofItaly

acceptedontherecommendationof

Prof.Dr.RolandW.Scholz,examiner

Prof.Dr.ThomasKöllner,co‐examiner

Prof.Dr.TimL.McDaniels,co‐examiner

2010

1

Summary

An increasing production of scientific literature addresses the past, current and future

importanceofecosystemsfortheprovisionofgoodsandservicestosociety.However,pressure

fromclimatechange,nationalpoliciesandglobalizationare increasinglydegradingthiscapacity

of ecosystems. Typically, the design of societal responses to ecosystem services degradation

requires considering the decision‐making ofmultiple actors from the local scalewhere ES are

providedtoregional,nationalandinternationalscaleswheredecisionsonrulesandresourcesare

defined and distributed. This is the case of SRS degradation which requires consideration of

decision processes also considering the on‐ and off‐site benefits of SRS. Indeed, soil erosion

affects upstream land users and downstream user as hydropower dam due to siltation.

Moreover, inawatershed,theseactorsdirectlybenefitingfromandmakingdecisionsregarding

SRSareembeddedindecisionprocesseshappeningatscalesthatgobeyondthewatershedand

beyondthedirectuseofSRSservices(e.g.,decisionstodesigntechnicalsolutionstosoilerosion,

decisionontheprovisionofincentivesandtheirtypes,etc.).

Thegoalofthethesisistoanalysethemulti‐hierarchyhumandecisionprocessescharacterizing

SRSprovisionconsideringthehumanandenvironmentsystemastwodifferent,complementary

andinterrelatedsystems.InthisthesisIanalysethecasestudyofSoilRegulationServices(SRS)in

theBirriswatershedCostaRica,acountrypioneerinpoliciestoprotectecosystemservices.The

Birris watershed is a relatively small territorywith an area of 5800 has. It is characterized by

intensiveandmarket‐orientedagriculturalproductionactivitieswithhighfragmentedfarmareas

andsteepslopes.Soilerosionhasbeenincreasingsinceseveraldecadesduetoinadequateland

use practices and increasing extreme precipitation events whose intensity and frequency is

expectedtogrowunderclimatechange.TheBirriswatershedrepresentsalearningcaseformany

nationalandregionalactorsduetoitscommonalitywithotherwatershedsintheregion.

Thefirstoffourarticlesofthisthesisanalyzestheenvironmentaldimensionfocussingtheroleof

forest ecosystem cover on hydrological responses. In the following three papers I address the

human dimension focusing on decision processes occurring at different scales: i) from the

individualfarmersdecision‐makingonsoilconservation;ii)tothewatershedscalewhereactors

aredirectlyaffectedbyon‐siteandoff‐siteprovisionofSRS;and,finally,iii)tothenationalcross‐

scale information‐sharing network of multiple actors involved directly or indirectly (scientists,

2

regulators, farmers’ associations, users of SRS, etc.) in themanagement ofwatersheds and of

SRS.

More specifically, the goal of the first paper is to synthesize the findings of several paired‐

catchmentexperimentsfromAfrica,AsiaandLatinAmericathatanalyzetheeffectofforestcover

changeonhydrologicalresponses.Weusemeta‐analysisasatoolused inecologicalstudiesto

synthesize results from different studies. This help us overcoming the small “n” problem

associated to many of these types of forest hydrology studies in watersheds especially in

developingworld.

The second paper addresses the complexity of farmers’ decision‐making in respect to soil

conservation decisions. This is analyzed through amulti‐dimensional decisionmodel including

economic cognitive and territorial variables influencing farmers’ decision‐making regarding soil

conservationefforts.Thestructuredsurveyincludeditemsbuiltthroughpreviousmeetingswith

farmerstocapturetheirperspectivesandunderstandingonsoilerosionanditssolutions.

The goal of the third paper is to analyze aspects that are relevant to build a collaborative

mechanismamongusersofon‐siteandoff‐siteprovisionofSRS.Specificmethodsfromdecision

scienceandnegotiationanalysisareappliedimplyingconsultationswithbothusersandproviders

ofSRS.Inseparatemeetings,consultationsallowedstructuringtheirfundamentalobjectivesand

identifyingkeyaspectsofcomposingadesiredmechanismsuchas: i)howtoselect farmers; ii)

what typeof contract should beused; iii)whoshould intermediateand, finally, iii) the typeof

possible incentives for farmers’soilconservation. Inorderto identifynegotiationspaceforkey

aspects of a mechanism, preferences of both stakeholders in respect to different alternative

aspectsofamechanismwereelicitedintwoseparatefocusgroups.

Thegoalofthefourthpaperistoidentifycross‐scaleinstitutionalmismatchesarisingfromformal

policiesandmandatesandconstrainingSRSprovisionanduse.Weuse“betweenesscentrality”

algorithm (common to social network science) to test how structural analysis of information‐

exchange network can identify boundary organizations which are potentially strategic to

overcomecross‐scaleinstitutionalmismatches.Weanalyseactors’officialmandatescontributing

directlyorindirectlytoSRSprovisionandtheirinteractionininformation‐sharingnetwork.

The analysis of the environmental dimension proves the usefulness and methodological

limitations of using meta‐analyses to synthesize findings from paired‐catchments experiments

3

studiesonthehydrologicaleffectsofchangesinnaturalandplantedforestscover.Overallresults

from experiments from Asia, Africa and Latin America show that forest cover can play an

importantroleindiminishingthebaseflowinwatershedbutitseffectonstorm‐flowcontrol(i.e.

water runoff causing erosion) depends more on local characteristics. Some methodological

limitationsfromthisuseofquantitativemeta‐analysiscanalsobeoutlined.Wefoundarelatively

smallnumberofpaired‐catchmentexperimentstudies fromAsia,AfricaandLatinAmericathus

limiting the capacity to analyze the interacting effect of important factors for water flow

regulation,suchassoil,geology,topography,orlandmanagementpractices.Asforthecapacity

of forests to control storm flows (related to increase in erosion) data found in the scientific

articlesusedinthemeta‐analysisdidnotallowaccountingfortheeffectsoffrequentandintense

extremeprecipitationevents.ThisalsolimitsthecapacitytocomparetheprovisionofSRSunder

climate change from natural forest ecosystem vs non‐forest land uses such as conservation

agriculture.

Atthelocalscale,farmers’awarenessoftheirexposureleveltosoilerosioncombineswithother

variables to determine their level of soil conservation efforts. The decision model includes

socioeconomic,territorialandcognitivevariablessuchasbeliefs,valuesandriskperceptionand

clearlyseparatesthreegroupsoffarmersbasedontheirsoilconservationefforts.Mostfarmers

areawareoftheriskoferosionalthoughsocioeconomicaspectssuchastypeofproductionand

farm size indicate that perceived opportunity cost given the farm production context might

hinder their conservation efforts. Farmers with low perception of erosion risk might also be

expressing “availability heuristic” paradigm due to their daily experience with erosion in the

watershed.

Atthewatershedscale,thedesignofcollaborativeeffortsfortheon‐andoff‐siteprovisionofSRS

requiresagreementonthefundamentalobjectivesofamechanismforcollaborativeefforts for

soilconservation.Consultedfarmersandhydropoweragreeontheimportanceofthepromotion

of learning through technical assistance andmonitoring of soil conservation programs and the

fairdistributionofincentives.Directpaymentforsoilconservationisonlylimitedlyconsideredas

a desired incentive alternative. Consistent with the fundamental goal of promoting learning,

technicalassistanceisseenasamoredesirablealternativethandirectpayments.

4

The national cross‐scale analysis of governance structure for SRS highlights that important

regulatorymismatches affect the definition of societal responses at the local level (i.e. where

directactionstopromoteadequateprovisionanduseofEShappen).Networkanalysishelpsus

identifying the information‐bridging characteristics of actors in informal information‐sharing

networks.Thisanalysisoutlinestheboundaryroleofthewatershedagricultural‐extensionoffice

helpingdiffusinginformationonimpactsaswellassocialandtechnicalfeasibilityofresponsesto

SRSdegradationacross‐scalesandpolicyareas.

Overall the thesis’ results show that soil conservation policies to support the provision of SRS

wouldbenefitfromtheuseofmixedpolicies.Thismightincludeprogramstoraiseawarenesson

currentand future soil erosion risks,promote learningamong farmers, and institutionalize the

boundary role of agricultural extension offices for their importance to promote learning and

adaptivemanagementofSRS.Thisisespeciallyvaluableinthecontextofareashighlyexposedto

increasingfrequencyofextremeprecipitationeventssuchasCentralAmerica.Moreover,inthe

faceofhighuncertaintiesandscarcityofdata(e.g.ontheimpactsoflanduse/managementand

climate change),mechanisms to update and disseminate information over timeon impacts on

soil erosion and correspondent solutions are required. In this respect, strengthening the

boundaryroleofagricultural‐extensionofficecanpotentiallyhelpupdatinginformationavailable

toscientists, regulatorsandfarmerson impactsandsocialandtechnical feasibilityofsolutions.

Thismightproveastrategytoaddresssomeoftheregulatorymismatchesthathinderresponses

toSRSdegradationatlocallevelandpromoteadaptivemanagementofsoilregulationservices.

6

Zusammenfassung

DiewissenschaftlicheLiteraturbefasstsichzunehmendmitdervergangenen,gegenwärtigenund

zukünftigen Bedeutung von Ökosystemen bei der Bereitstellung von für die Gesellschaft

essentiellen Gütern und Dienstleistungen. Diese Leistung von Ökosystemen ist jedoch

zunehmend gefährdet durch Klimawandel, nationale Politik und Globalisierung. Für die

Gestaltung von gesellschaftlichen Massnahmen gegen die Degradierung von

Ökosystemdienstleistungen müssen die Entscheidungsprozesse verschiedener Akteure

berücksichtigt werden: von der lokalen Ebene, auf der Ökosystemdienstleistungen geleistet

werden, über die regionale, nationale und internationale Ebene, auf der Entscheidungen über

Regeln und Ressourcenverteilung gefällt werden. Die Entscheidungsprozesse internationaler

Akteure müssen auch bei der Degradierung von Bodenregulierungsdienstleistungen

berücksichtigt werden, und systeminterne und ‐externe Nutzen berücksichtigt werden.

Bodenerosion beeinflusst flussaufwärts die Landnutzer und flussabwärts die Betreiber von

Staudämmen durch zunehmende Verschlammung. In einem Wassereinzugsgebiet sind die

Akteure, welche Entscheidungen zu Bodenregulierungsdienstleistungen fällen und davon

profitieren auch in Entscheidungsprozesse involviert, die über das Einzugsgebiet‐ und über die

direkte Nutzung von Bodenregulierungsdienstleistungen hinausgehen (z.B. Entscheidungen zu

technischen Lösungen für Bodenerosion oder Entscheidungen zur Bereitstellung von Anreizen

undderenAusgestaltung,etc.).

Das Ziel dieser Arbeit ist es die multi‐hierarchischen Entscheidungsprozesse, welche die

Bereitstellung von Bodenregulierungsdienstleistungen charakterisieren, zu analysieren. Dabei

werdendieMensch‐Umwelt‐Systemealszweiunterschiedliche,komplementäreundverknüpfte

Systeme verstanden. In dieser Arbeit untersuche ich die Bodenregulierungsdienstleistungen in

einerFallstudieimGebietdesBirri‐WassereinzugsgebietinCostaRica‐einemLandwelcheseine

Pionierrolle übernimmt in der Politik zur Erhaltung von Ökosystemdienstleistungen. Das Birri‐

EinzugsgebietistrelativkleinmiteinerFlächevon5800Hektaren.Charakterisiertwirdesdurch

intensive undmarktorientierte Landwirtschaft, stark fragmentierte Landwirtschaftsflächen und

steile Hänge. Die Bodenerosion hat in den letzten Jahrzehnten stark zugenommen. Gründe

hierfür sind nicht‐angepasste Landnutzungspraktiken und zunehmende Niederschläge, deren

Intensität und Häufigkeit mit wandelndem Klima weiter ansteigen werden. Das Birri‐

7

Einzugsgebiet dient als Lernbeispiel für nationale und internationale Akteure, da es viele

GemeinsamkeitenmitanderenEinzugsgebietenaufweist.

DererstevonvierArtikelndieserArbeitanalysiertdieUmweltdimensionundbehandeltdieRolle

der Ausdehnung von Waldökosystemen auf hydrologische Reaktionen. In den folgenden drei

Artikeln befasse ich mich mit der menschlichen Dimension und fokussiere auf

EntscheidungsprozesseaufverschiedenenEbenen:i)vomindividuellenEntscheideinesBauernzu

Bodenerhaltung; ii) zur EbenedesWassereinzugsgebiets,woAkteuredirektbetroffen sindvon

systeminterner und ‐externer Bereitstellung von Bodenregulierungsdienstleistungen; und

schliesslich iii) zum nationalen Netzwerk, welches ebenenübergreifenden zwischen

verschiedenenAkteurenInformationenaustauscht,welchedirektoderindirektimManagement

des Einzugsgebiets involviert sind (wie beispielsweise Wissenschaftler, Behörden,

Bauernverbände,NutzerderBodenregulierungsdienstleistungen).

Spezifisches Ziel des ersten Artikels ist es die Resultate von Experimenten mit gepaarten

Einzugsgebieten inAfrika, Asien und Lateinamerika zusammenzufassen,welche den Effekt von

veränderterWaldausdehnungaufhydrologischeReaktionenanalysieren.MittelsMeta‐Analysen

werden die Resultate verschiedener Studien zusammengeführt und das Problem vom kleinem

“n” bewältigt, welches bei vielen Studien zur Waldhydrologie, insbesondere in

Entwicklungsländern,auftritt.

Die zweitePublikationbefasst sichmitderKomplexitätderEntscheidungsprozessevonBauern

bezüglich Bodenerhaltung. Dies wird anhand eines multi‐dimensionalen Entscheidungsmodells

analysiert, welches ökonomisch kognitive und territoriale Variabeln beinhaltet, welche die

Entscheidungsprozesse von Bauern bezüglich Bodenerhaltungsanstrengungen beeinflussen. Die

strukturierteUmfrage beinhaltet Elemente aus vorgängigen Treffenmit Bauern, bei denen die

AnsichtenunddasVerständnisvonBodenerosionundderenLösungenuntersuchtwurden.

DasZieldesdrittenArtikelsistdieAnalysevonAspekten,dierelevantsindfürdenAufbaueines

gemeinschaftlichen Mechanismus unter systeminternen und ‐externen Nutzern von

Bodenregulierungsdienstleistungen. Spezifische Methoden der Entscheidungsforschung und

Verhandlungsanalysewerdenangewandt,welcheBefragungenvonNutzernundAnbieternvon

Bodenregulierungsdienstleistungen beinhalten. Getrennte Konsultation von Akteuren

ermöglichten es die fundamentalen Ziele zu strukturieren undHauptaspekte von gewünschte

8

Mechanismenzuidentifizieren,wiebeispielsweise:i)wiesollendieBauernausgewähltwerden;

ii) welche Kontakttypen sollten verwendet werden; iii) wer sollte Konflikte schlichten und

schliesslich iv) welche Arten von Anreizen gibt es für Bauern für die Umsetzung von

Bodenerhaltungsmassnahmen. Um die Verhandlungsspielräume für die Hauptaspekte des

Mechanismus zu erkennen wurde mit jeder Akteursgruppe eine separate Fokusgruppe

durchgeführt und die Präferenzen der Akteure zu verschiedenen alternativen Aspekten eines

Mechanismuseruiert.

Das Ziel der vierten Publikation ist es ebenenübergreifende institutionelle Diskrepanzen zu

identifizieren, welche aus offiziellen Regeln und Mandaten entstehen und welche die

BereitstellungundNutzungvonBodenregulierungsdienstleistungeneinschränken.Wirbenutzen

“betweeness centrality” Algorithmen (welche häufig in sozialer Netzwerkanalyse verwendet

werden) um zu testen wie die strukturierte Analyse des Informationsaustauschs‐Netzwerks

Schlüsselorganisationen identifizieren kann, die potenziell eine strategische Position innehaben

um ebenenübergreifende institutionelle Diskrepanzen zu bewältigen. Wir analysieren die

offiziellen Mandate der Akteure, welche direkt oder indirekt zur Bereitstellung von

Bodenregulierungsdienstleistungen beitragen und untersuchen deren Interaktion in

Informationsaustausch‐Netzwerken.

Die Analyse der Umweltdimension zeigt den Nutzen und die methodologischen Grenzen von

Meta‐Analysen für die Synthese von Resultaten aus hydrologischen Experimenten, in unserem

Fall Experimente mit gepaarten Einzugsgebieten, bei denen die hydrologischen Effekte von

veränderterFlächevonnatürlichenundgepflanztenWäldernanalysiertwerden.Gesamtresultate

ausAsien,AfrikaundLateinamerikazeigen,dassdieWaldbedeckungeinewichtigeRollespieltbei

derVerringerungdesBasisabflussesimWassereinzugsgebiet,aberderenEffektaufdieKontrolle

des Spitzendurchflusses (d.h. erosionsbedingter Wasserabfluss) hängt mehr von lokalen

Charakteristikenab.EinigemethodologischeGrenzenvonquantitativenMeta‐Analysenkönnen

ebenfalls umrissen werden. Wir fanden eine relativ kleine Anzahl Studien mit gepaarten

Einzugsgebiets‐Experimenten aus Asien, Afrika und Lateinamerika. Dies schränkt die

Analysemöglichkeiten von interagierenden Effekten wichtiger Wasserflussregulationsfaktoren,

wie Boden, Geologie, Topographie oder Landnutzungspraktiken, ein. Die in der Meta‐Analyse

von wissenschaftlichen Artikeln gefunden Daten zur Kapazität von Wäldern Sturmflüsse zu

9

kontrollieren (welche zu erhöhter Erosion führen) erlaubten es nicht, die Effekte häufiger und

intensiverExtremniederschlägezuberücksichtigen.DiesschränktauchdieVergleichsmöglichkeit

von der Bereitstellung von Bodenregulierungsdienstleistungen unter Klimawandel durch

natürliche Waldökosysteme und nicht‐Wald Landnutzungen wie beispielsweise Conservation

Agriculture(LandwirtschaftmitkonservierenderBodenbearbeitung),ein.

AuflokalerEbenesinddieAnstrengungenderBauernumdenBodenzuerhaltenunteranderem

bestimmt von ihrem Bewusstsein wie fest sie der Bodenerosion ausgesetzt sind. Das

Entscheidungsmodell beinhaltet sozio‐ökonomische, territoriale und kognitive Variablen wie

Überzeugungen,WerteundRisikowahrnehmungundunterscheidetzwischendreiGruppenvon

Landwirten, je nach Anstrengungen zur Bodenerhaltung. Die meisten Landwirte sind sich des

Risikos der Erosion bewusst, obwohl sozio‐ökonomische Aspekte, wie Produktionstypen und

Hofgrössedarauf hindeuten, dasswahrgenommeneundkontextgegebeneOpportunitätskosten

die Anstrengungen zum Bodenschutz eindämmen können. Bauernmit geringerWahrnehmung

desErosionsrisikoskönntenebenfallsdasParadigmaderVerfügbarkeitsheuristik zumAusdruck

bringen,dasietäglichErfahrungenmitErosionimWassereinzugsgebietmachen.

Auf der Ebene des Wassereinzugsgebiets erfordert der Aufbau von gemeinschaftlichen

Anstrengungen für eine systemsinterne und ‐externe Erbringung der

Bodenregulierungsdienstleistungen die Vereinbarung von grundsätzlichen Zielen eines

Mechanismus’fürgemeinschaflicheLeistungenzuGunstenderBodenerhaltung.BefragteBauern

und Wasserkraftakteure sind sich einig über die Wichtigkeit von Weiterbildung und fairer

VerteilungvonAnreizen.DirektzahlungenfürBodenerhaltungwirdnurineinemkleinerenMasse

als erwünschte Alternative betrachtet. Technische Unterstützung ist mehr erwünscht als

Direktzahlungen,wasmitdemGrundzielderFörderungvonWeiterbildungübereinstimmt.

Die nationale ebenenübergreifende Analyse von Lenkungsstrukturen für

Bodenregulierungsdienstleistungen betont, dass wichtige regulatorische Diskrepanzen lokale

gesellschaftliche Reaktionen beeinflussen (d.h. da, wo direkte Handlungen zur ausreichenden

BereitstellungunddieNutzungvonÖkosystems‐Dienstleistungenstattfinden).

Mittels Netzwerkanalyse können wir die informationsüberbrückenden Charakteristiken von

AkteurenininformellenInformationsaustausch‐Netzwerkenidentifizieren.DieseAnalysestreicht

10

dieSchlüsselrolle von landwirtschaftlichenBeratungsbüros imEinzugsgebiethervor,welchebei

der Verbreitung von Informationen zu Umwelteinflüssen und technischer Umsetzbarkeit von

Bodenschutzmassnahmen helfen, und dabei Informationen über Politikbereiche und Ebenen

hinausaustauschen.

GesamthaftzeigendieResultatedieserArbeit,dasspolitischeMassnahmenzurBodenerhaltung

zur Unterstützung von Bodenregulierungsdienstleistungen, von einer Kombination von

verschiedenenMethoden profitieren würde. Dies könnte Programme beinhalten, welche das

Risikobewusstsein von heutiger und künftiger Bodenerosion stärken, die Weiterbildung von

Bauern fördern und die Institutionalisierung der Schlüsselrolle der landwirtschaftlichen

Beratungsdienste, welche für die Förderung von Weiterbildung und angepasstem

Bodenmanagementswichtig sind, vorantreiben.Dies istbesonderswichtig inRegionen,welche

stark betroffen sind von zunehmenden Extremniederschlägen, wie beispielsweise

Zentralamerika. Angesichts grosser Unsicherheiten und Datenmangel (z.B. zum Einfluss von

Landnutzungund‐managementundKlimawandel),werdenMechanismenzurAufdatierungund

kontinuierlichen Verbreitung von Informationen zu Einflüssen auf Bodenerosion, sowie den

entsprechenden Lösungen, benötigt. Die Stärkung der Schlüsselrolle der landwirtschaftlichen

Beratungsdienste kann hier einen potenziellen Beitrag leisten zur Aufdatierung von

InformationenzuEinflüssenundsozialerundtechnischerMachbarkeitvonLösungen,umdiese

fürWissenschaftler,RegulatorenundBauernverfügbar zumachen.Diesehabeneinen Einfluss

auf die soziale und technischeUmsetzbarkeit der Lösungsvorschläge.Dies könnte sich als eine

Strategiebeweisen,umgewisseregulatorischeDiskrepanzenanzugehen,welchedieReaktionauf

dieDegradierungvonBodenregulierungsdienstleistungenauflokalerEbeneverhindern,undum

einangepasstesManagementvonBodenregulationsdienstleistungenzufördern.

12

TableofContents

Summary_____________________________________________________________ 1Zusammenfassung_____________________________________________________ 61. Introduction _______________________________________________________ 16

1. 1 Relevance of soil erosion ___________________________________________ 16

1.2 On- and off-site benefits of Soil Regulation Services (SRS)________________ 17

1.3 Soil erosion control: a human-environment paradigm ___________________ 19

1.4 Purpose and research questions of the thesis ___________________________ 22

1.5 Meta-analysis of forest-cover change effects on water-related ecosystem services _____________________________________________________________ 23

1.6 Farmers’ decision model regarding soil conservation effort _______________ 23

1.7 Two-party negotiation analysis for a mechanism to promote Soil Regulation Services _____________________________________________________________ 24

1.8 Institutional mismatches and information-sharing in multi-actors’ and cross-scale governance for Soil Regulation Services _____________________________ 24

1.9 Goals and justification of the methods used ____________________________ 25

1.10 References_______________________________________________________ 27

2. Managing watershed services of tropical forests and plantations: can meta-analyses help?________________________________________________________ 30

2.1 Introduction ______________________________________________________ 30

2.2 Method __________________________________________________________ 32

2.3 Results and discussion ______________________________________________ 342.3.1 Selected studies _______________________________________________ 342.3.2 Natural forests________________________________________________ 372.3.3 Planted forests ________________________________________________ 402.3.4 Usefulness of the meta-analyses __________________________________ 402.3.5 Limitations ___________________________________________________ 42

2.4 Conclusions ______________________________________________________ 42

2.5 References________________________________________________________ 43

3. Decision making by farmers regarding ecosystem services: factors affecting soil conservation efforts in Costa Rica________________________________________ 49

3.1 Introduction ______________________________________________________ 49

3.2 Farmers’ Decision-Making for Soil Conservation _______________________ 51

3.3 Methods _________________________________________________________ 543.3.1 Description of the area __________________________________________ 543.3.2 Sample and Survey Instrument Design _____________________________ 553.3.3 The Decision Model____________________________________________ 55

13

3.3.4 Statistical analysis _____________________________________________ 613.3.4.1 Construction of cognitive variables ____________________________ 61

3.3.4.2 Differentiation of farmers’ types_________________________________ 63

3.4 Results___________________________________________________________ 643.4.1 Sample description_____________________________________________ 643.4.3 Differentiation of farmers’ groups _________________________________ 663.4.4 Testing hypotheses of differences between farmers’ groups _____________ 67

3.5 Discussion ________________________________________________________ 70

3.6 Conclusions ______________________________________________________ 72

3.7 References________________________________________________________ 72

4.1 Negotiation analysis for mechanisms to deliver ecosystem services: a value-focused bargaining structure to achieve joint gains in the provision of Soil Regulation Services _____________________________________________________________ 77

4.1 Introduction ______________________________________________________ 77

4.2 Negotiation Analysis and Its Relevance for ES __________________________ 80

4.3 ES provision in the Birris Watershed _________________________________ 82

4.4 Methods _________________________________________________________ 844.4.1 Data Acquisition (Step 1)________________________________________ 864.4.1.1 Data from farmers ____________________________________________ 864.4.1.2 Data from JASEC ____________________________________________ 874.4.2 Synthesis for template building (Step 2) ____________________________ 884.4.3 Verification and evaluation of alternative strategies (Step 3) ____________ 88

4.5 Results___________________________________________________________ 894.5.1 Objectives of negotiating parties __________________________________ 894.5.1.1 Farmers ____________________________________________________ 894.5.1.2 JASEC_____________________________________________________ 914.5.2 Alternatives as strategies across several dimensions ___________________ 924.5.3 Acceptability, preferred alternatives and zones of bargaining ____________ 97

4.6 Discussion ________________________________________________________ 99

4.7 Conclusions _____________________________________________________ 102

4.8 References_______________________________________________________ 103

5. Governance across-scales: regulatory mismatches and the role of boundary organizations for soil regulation-services under climate change in Costa Rica ___ 108

5.1 Introduction _____________________________________________________ 109

5.2 Concepts ________________________________________________________ 1115.2.1 Information and knowledge exchange in ES governance structures ______ 1115.2.2 Multiple scales and regulatory mismatches for SRS __________________ 1125.2.3 Boundary organizations for ES governance networks _________________ 113

5.3. Context for the Birris case study____________________________________ 114

5.4 Methods ________________________________________________________ 115

14

5.4.2 Network analysis _____________________________________________ 1165.4.2.1 Building the questionnaire for network analysis____________________ 1175.4.2.2 Matrix analysis of network data ________________________________ 118

5.6 Results__________________________________________________________ 1205.6.1 Network boundary, formal policies and mandates of organizations across scales___________________________________________________________ 1205.6.2 Structural analysis of the governance network_______________________ 121

5.7 Discussion _______________________________________________________ 125

5.8 Conclusions _____________________________________________________ 131

5.9 References_______________________________________________________ 131

6. Conclusions ______________________________________________________ 136

6.1 Research questions and their rationales ______________________________ 136

6.2 Unanswered research questions, limitations and future research _________ 138

6.3 Main findings ____________________________________________________ 140

6.4 Lessons learned for other cases on ES degradation _____________________ 141

6.5 __________________________________________________________Key messages___________________________________________________________________ 144

7.Annexes __________________________________________________________ 146Annex 1: Extended acronyms of organizations in the network and their specific mandates___________________________________________________________ 147Annex 2. Ecosystem-based adaptation to climate change: what role for policy-makers, society and scientists?_________________________________________________ 163

A 2.1 Introduction ___________________________________________________ 163

A 2.2 Content of the workshop _________________________________________ 164

A 2.3 Key messages for national policy-makers ___________________________ 166

A 2.4 Key messages for communities, local actors and others members of civil society _____________________________________________________________ 167

A 2.5 Key messages for scientists _______________________________________ 168

A 2.6 Conclusions____________________________________________________ 170

Acknowledgments____________________________________________________ 172Curriculum Vitae ____________________________________________________ 174

16

1. Introduction

1.1Relevanceofsoilerosion

Timeneeded to formsoils ishighlydifficult to determinatealthough studiesaround theworld

suggestthat,dependingonspecificinteractionsamongclimate,humanandgeologicalprocesses,

uptohundredsofyearsmightberequiredtodevelopsoils(Chesworth,2008).Soilfertilityaswell

as itsdegradationduetoerosionhavebeenastrongdriverofcorrespondinglycivilizationraise

anddecline(McNeillandWiniwarter,2004).Currently,onlylessthanonefifthoftheearth’sland

ispotentiallyproductiveandavailabletoprovide97percentoffoodsupply(therestfromwater

systems).However, future food supply is bound to expand the intensification and frontiers of

agriculture increasingerosionespecially inmarginal lands(Morgan,2005).Evidenceforthisare

suggestedbythecurrentratesoferosioninagriculturallandsthatarebylarge(i.e.between13

and40times,PimentelandKounang,1998)greaterthannaturalsoilproduction,soil lossunder

naturalecosystemsandlong‐term‐geologicalerosion(Montgomery,2007).

Highratesoferosion(from20to100t/ha/yr)areproducingadeclineinproductivityasmuchas

15to30percentannuallywhichcorresponds(whenconsideringthedifficultyofrestoringtheir

fertility) to a loss of 6million hectares annually (Pimentel et al., 1995). Around 85%of global

agricultural land is affected by some form of degradation including erosion, salinization and

compaction(WRI,2000).AccordingtoUNEP’sGlobalAssessmentofSoilDegradation(GLASOD),

soilerosionaccountsfor82percentofthehuman‐inducedlanddegradationthusaffectingmore

thanabillionhectares,althoughonly0.5%ofwhichhasreachedanirreversiblestage(Oldeman,

1992). These figures bear inherent uncertainty in their estimation formethodological reasons

such as scaling up values from local assessments to large regional studies or using national

evaluation standards instead of a unique global one.However, even considering uncertainties,

they show that agricultural practices can reduce significantly soil erosion and provide food

security as well as intermediate services to different sectors of society (e.g. water users,

hydropowerdams,carbonmarkets)(Morgan,2005;Fisheretal.,2009).

In Latin America a major cause of soil degradation is associated to inappropriate soil

management practices in agricultural land and in Mesoamerican Isthmus reaches 74% of

croplands(SantibaniezySantibaniez,2007).Marginalityofsmall landowners insteep landsand

17

intensificationofagriculturalproductionareimportantdriversofcurrentsoilerosion.Observed

andprojectedfrequenciesofextremeprecipitationeventsinMesoamerica(Aguilaretal.,2005)

indicatethatclimatechangeandvariabilityareexpectedtoexacerbatesoilerosionandfurther

affect soil productivity in agricultural lands. For this challenging contexts, Santibaniez and

Santibaniez (2007) indicatethatamongtheseveraladaptationoptionstoreducesoilerosion in

agriculturallandsofLatinAmericaimprovedinformation‐sharingsystemsandinnovativefinancial

mechanismsareimportant.Theformerarenecessaryforearlywarningsagainstextremeclimate

aswellas for increasingfarmers’knowledgeon the implementationofsoundsoilmanagement

practices (Cash, 2001). Innovative financial mechanisms are also important to increase the

capacitytoaccesstechnologiesandinputsforbettersoilmanagement(Pagiola,2008).

1.2On‐andoff‐sitebenefitsofSoilRegulationServices(SRS)

The combined effect of upstream farmers’ decision regarding soil management and the

increasingoccurrenceofextremeprecipitationeventsdeterminethelevelofSRSprovisionboth

on‐site and off‐site. In this section I discuss the specific interaction of human and natural

componentsdefiningthedegradationofSRSasaresultofsoilerosion.

Morespecifically,soilerosioncanbedefinedasprocessincludingadetachmentofsoilparticles

andtheirtransportbywaterorwinduntilenergyisnotsufficientanddepositionoccurs(Morgan,

2005).When considering theerosioncontrol capacityofnaturalecosystems, it is important to

distinguishandtoestablishtheirrelative importanceandsite‐specificcontributionstodifferent

typesoferosion.Indeed,naturalforestecosystemstendtosufferrelativelylittlesurfaceerosion

(ChomitzandKumari,1996)whiletheycancontributelimitedlytocontrolsoilmasserosion(i.e.

deeperthan3m)or intherecoveryofsoilstructurefromprolongedexposuretogullyerosion.

Throughoutthisthesisweusetheterm“soilerosion”torefertosurfaceerosionasproducedby

rain splash, overland flow and rill erosion which are influenced by soil, water and vegetation

managementpracticesfarmerscandecidetoimplementintheirfarms(Morgan,2005).

18

Thefactorscontrollingsoilerosionaretheerosivityoftheerodingagent(e.g.precipitationand

wind), the erodibility of the soil (i.e. function of soil types1), the slope of the land (e.g. its

steepness, itsshapeand its length)andthenatureoftheplantcoverandfinallytheenergyof

precipitationdrops(Morgan,2005).Thesefactorsinteractwitheachotherinacomplexmanner.

For example, soil erodibility depends on interaction between geological and geomorphologic

processeswithsitespecificfloraandfaunaoverhundredsofyears(Chesworth,2008)whilethe

effectofprecipitationandofvegetationcovercanhaveimmediateeffects(e.g.densevegetation

cover can reduce soil erosion). Current societal decisions to define responses depend on

awarenessofthesecomplexcausalloopsandonthecontingentaction‐contextoferodingagents.

Forexample,precipitationerosivity(themostimportantsoilparticles‐detachingagent;Morgan,

2005) ishigherwhenbareunprotected soil surface isexposed to rain splashor runoff laminar

erosion(WischmeierandSmith,1978).Farmlevelpossibleresponsestosoillaminarerosioncan

includemeasuresuchasmanagementofvegetationcover,infiltrationcapacityandsurfacerunoff

kineticpotentialormanagementoforganicmatter fostering soil structuration (Morgan,2005).

Society ideally should aim at achieving a rate of erosion control that equals that of new soil

formationalthoughwherethisbalanceliesforspecificsitesisunknown.Currentannualaverage

soil regeneration rate of world are 0.1mm*yr‐1 while soil erosion rates in some areas of the

worldareover10mm*yr‐1 (Morgan,2005).Farmers’alternativestoreducesoilerosion include

measures to contrast detachment phase of soil particles such as agronomic measures (e.g.

managingvegetationcover)andsoilmanagementpractices(e.g.ploughingwithlowimpactson

soilstructure).Mechanicmethods(e.g.engineeringmanagementofslopetopographyoruseof

syntheticpolymers,Ortsetal.,2007)areinsteaddirectedtocontrolthetransportphaseofsoil

particles.Agronomicmeasuresarethemostcost‐effectiveandsuitabletomostfarmingsystems

(Morgan,2005).

Thescientificobservationoferosionaswellasitsmanagementmusttakeintoaccountwhether

both the on‐site and off‐site effects are considered or only one of them due to time‐lags

characterizingthetransportofparticlesdownstream.Indeed,whileon‐siteeffectsoferosioncan

be both visible and perceivable in short‐time horizon, its off‐site effects on sedimentation

1 For example, volcanic soils such as Andosols are prone to heavy erosion in Central America since they loose physicalstructuresduetotheirrecentformation.

19

downstreammustconsiderthecharacteristictime‐lagbetweenon‐siteremovalofasoilparticle

anditsoff‐siteobservationwhichtendsto increase(duetoincreaseofsedimentstorageareas)

with the dimension of the catchment (Schoenholtz, 2004; Riedelet al., 2004). In specific sites

wherenatural sedimentproduction is lowdue toon‐sitenaturalequilibrium, forest clearing is

bound to contribute to more erosion. Indeed, site‐ and species‐specific relationship are

importantandnatural forestshouldbeseenasthenaturalbaseline forerosioncontrolagainst

whichallotherlandusesshouldbecompared(Calder,2002).

1.3Soilerosioncontrol:ahuman‐environmentparadigm

Theanalysisofsoilerosioncontrolrequiresconsideringcomplexityofactorsandtheirdecision‐

making, spatial and temporal scales and the interactions/feedbacks from action on the

environment(i.e.soil) to itsconsequencesonhumanactivities.Sinceourgoal is tounderstand

how human (defined as the social system ranging from individuals to society) interacts with

environmentalsystemandwhattransitionpathwaycanbeoutlinedtoimprovemanagementof

soil resources, we can refer to the Human‐Environment System (HES) paradigm (Scholz and

Binder, 2004) as a framework allowing us to analyse this complex problem under a human‐

systemperspective(i.e.tounderstandimpactsofregulatorymechanismsinsideamulti‐hierarchy

human system). Under this perspective, soil erosion control can be defined based on the

decision‐making that society needs to consider for their management (Fisher et al., 2009).

Accordingtotheseauthors,classificationofecosystemservicessuchasSoilRegulationServices

must take into account the linkages among intermediate services such as soil formation, final

services such as fertility provision and actual benefits which accrue when the beneficiaries

actuallyisabletousetheseservicestoachievetheirgoals(e.g.producingagriculturalproducts).

The final achievement of beneficiaries’ intended goals using ecosystem services depends on

othertypeoffactorswhichconditiontheeffectiveuseoffinalservices2.

2 For example, hydropower production requires some built capital to harness thebenefitsofwaterregulationfromupstreamecosystems

20

Figure 1: Human-Environment System for soil erosion control in the Birris case study.

Then, under the decision‐making perspective to study the provision and the use of ecosystem

servicessuchassoilerosioncontrolwemustconsideron‐siteandoff‐siteprovisionofthisservice

and the corresponding actors interacting with the provision and/or use of this services. For

example, on‐site we must identify different actors/sectors playing different roles in the

management and use of soil erosion control services according towhether they influence the

provisionof intermediateservices(e.g.farmers influencetheformationofsoilswhilemanaging

organicmatter), the final service (e.g.which results from the interaction of soil characteristics

and the specific on‐site agricultural production type) and the actual benefit of producing

agriculturalproducts(e.g.whichmightbeinfluencesbyfarmers’accesstootherinputs).Onthe

otherside,off‐sitebenefitsofsoilerosioncontrolservicesdependonthecharacteristicsofthe

watershed dynamic transport of sediments downstream, on the specific requirements (e.g.

hydropower dams needs low sediment load water) and capacity of a user to concretize its

benefits.

21

Theseon‐siteandoff‐siteactorsmakedecisionsbasedontheirgoalsbutalsoontheiraccessto

resourcescomingfromotheractorsthatprovideinformationonerosioncontrolalternatives,on

laws defining incentives for improved soilmanagement, on the potential future threats to soil

erosion,etc. Alongwiththegeneral frameworkofHES, forthecaseofsoilerosioncontrolwe

needtoconsiderahierarchyofhumansystemsfromtheindividualfarmersconcernedbasically

withon‐siteaccesstoSRSbenefits, thedownstreamusersofoff‐siteSRSbenefits, theirmutual

interest in the provision of SRS and the regulatory frameworks in which each of them is

embedded as to guide legally their action, providing them with relevant information on

alternativeresponsestoSRSservicesdegradation.

Thus, goal formation and subsequent strategy formation to guide the transition (i.e. based on

actors’ values; Keeney, 1992) depend on capability, experiences and constraints of individuals

involved directly (e.g. farmers and hydropower downstream users of SRS) or indirectly (e.g.

regulators, scientistsproviding informationonalternative technically‐feasible responses) in the

problem of SRS degradation. Environmental awareness (i.e. of impacts, of externalities)

representstheinterfaceofthehuman‐environmentsystem ( )H E! whereindividualsaccording

totheirmentalmodeloftheenvironmentdefinewhataredirectandshort‐termconsequencesof

theiractionsontheenvironmentandwhataredelayedeffects.Inthespecificcaseofsoilerosion

control services farmers and downstream SRS users normally have a direct a short‐term

experience of environmental consequences of soilmanagement actions.However, the specific

environmentalcontextofSRS intrinsically includes longer‐termenvironmentalconsequencesof

humanactions.Forexample,soil formationinvolcanicsoilsmightprovidedeepsoilswherethe

effects of top‐soil losses due to erosionmight not be visible or experienced in the short‐term

productivecycleofthedirectusersofSRS. Itmightrequire longertimeuntil theconsequences

are evident. Similarly, climate change and variability can worsen the environmental

consequences of soil management actions in the longer term which might not be evident or

consideredindaytodaydecision‐makingofactors(SivakumarandNdiangui,2007).Inthiscase,

actorsdependalsoonaccesstoscientificinformationthatimprovestheirunderstandingofshort

and longer‐term consequences of their actions as well on the design of alternative regulatory

thatallowinvolvedactorstorespondandadaptovertimetochangingconditions(DuitandGalaz,

2008).TheproposedHES(ScholzandBinder,2004)providesaninterestingframeworksignalling

22

thattheenvironmentalconsequencesofhumanactionatonescale(e.g.atthefarmlevel)need

supportiveactionsatascaledifferentfromthatoftheaction(e.g.atthenationalorwatershed

scale designing new research on technology options and technology‐transfer mechanisms,

designingnewregulatoryincentivesforsustainablesoilmanagement,etc.).

Inthisrespect,awarenessofenvironmentalconsequencesofhumanactioninthecontextofSRS

services is relevant not only among those who are directly dealing with SRS degradation (i.e.

providers and users of SRS) but also among those who influence regulations, incentives and

scientific understanding on SRS. Alongwith the HES paradigm, these actors’ awareness of SEC

degradation and/or of their corresponding role potentially influence their decision‐making and

canthusconditiontheconstraintsandopportunitiesthatdirectly‐interestedactorsfacetocope

withthisenvironmentalproblem.ThisspecificprincipleoftheHESparadigmisespeciallyrelevant

since the complexity and deep‐uncertainty (Hulme, 2005) of the combined effects of climate

change and soilmanagement practices on SRS need updating information over time to adjust

actions undertaken. In otherwords, regulatory, incentive and science productionmechanisms

that allow human system to establish adaptive cycles are extremely important to cope with

changingconditions(i.e.changeinfrequencyandintensityofextremeprecipitationevents)that

arebyalargepartunknown(Cashetal.,2003;DuitandGalaz,2008).

1.4Purposeandresearchquestionsofthethesis

The objective of this study is to contribute to understanding of the human dimension of soil

degradationproblemusingamulti‐scaleapproach.Ourresearchquestionsarisefromtheanalysis

ofthefactorsthatcharacterizesoilerosioncontrol.Wethusconsidertheenvironmentalsystem

by analysing multiple studies on forest cover change‐water services relationship (i.e. through

meta‐analysis.Onthehumandimensionwetake intoaccountthemulti‐scalecharacteristicsof

soilerosionbyapproachingthedifferentdecisioncontextsthatateachscalearerelevant.Inthis

respect,farmers’decision‐makingregardingtheirfarm‐soilsaccountsfordecisionsthathavethe

most direct impacts of humans on the soil resource. At the watershed scale and using an

ecosystemserviceapproachweanalyse theaspects thatupstreamprovidersof SoilRegulation

Services(i.e.farmers)anddownstreamusersofSRS(i.e.hydropowerdam)considerimportantto

23

establishacollaborativeeffort(i.e.throughnegotiationanalysis)toimproveon‐siteandoff‐site

SRSprovision.Finally,weanalysethecross‐scaleinteractionofactorsfromdifferentpolicyareas

(e.g. regulators, scientists, NGOs, farmers’ associations, etc.) in information‐exchange and

multipleformal‐policyobjectivescontexts.Anadditionalarticlehasbeen includedtoreportthe

perspective of more than 80 experts on ecosystem‐based adaptation to climate change from

differentdisciplines inLatinAmerica.Keymessagesfromtheseexpertssupportthe importance

ofcommunicationandinformation‐sharingmechanismsaswelloutlinetheneedforintermediary

organizationbridgingdifferentknowledge‐communitiesfromthelocaltonationalscale.

1.5Meta‐analysisofforest‐coverchangeeffectsonwater‐relatedecosystemservices

Societyhas,sincelong,takenforgrantedthebeneficiaryroleofforestcoverinprovidingwater‐

related ecosystem services (Andreassian, 2004). General rule of thumb on the role of forest

ecosystems inprovidingwater‐relatedecosystemservices ishinderedbyhighsite‐specificityof

thisrelationship.Moreimportantforthethesis’focusonSRS,itisimportanttolookattheeffect

offorestcoveronhydrologicalprocessessuchasstormflowthatcancausesoilerosion.

Weusemeta‐analysis (Gurevitchetal., 2001) to runaquantitative synthesisof the findingsof

studies from different sites around the world. The limited number of studies on the specific

relationforestcover‐waterrunoffanderosionhinderthebenefitsofusingmeta‐analysistodraw

generalconclusions(Ilstedtetal.,2007).Wethusconcentrateonstudiesthatincludeanalysisof

the effects of forest cover on hydrological responses such as storm flow (defined as a part of

waterrunoff;UNESCO‐WMO,1992)whichisaproxyofsoilerosion(Bonell,1998).

Researchquestion:Canquantitativemeta‐analysishelpresolvethesmall“n”problemaffecting

decisions regarding the role of forest ecosystems in providingwater‐related services? Can this

methodhelpmakinggeneralconclusionsandsoguidepolicymakingonsoilcovermanagement?

1.6Farmers’decisionmodelregardingsoilconservationeffort

Many studies of adoption of control erosion measures by farmers have been based on

assumptions derived from neo‐classical economics (Van de Bergh et al., 2000). However,

evidences suggest that decision making of farmers is a complex combination of site‐

characteristicsandothercognitiveaspects.Thisisespeciallyrelevantconsideringthatobservable

24

effectsofsoilerosionarenotperceivedequallyamongfarmersandamonghigherandlowersoil

erosionriskareas.

Research question: Do risk perception, values and beliefs influence farmers’ soil conservation

behaviouroritisonlyeconomicreasoningthatguidefarmers’behaviour?

1.7Two‐partynegotiationanalysisforamechanismtopromoteSoilRegulationServices

Collaborative planning for complex human‐natural systems necessarily relies on negotiation

(InnesandBooher,1999);yethowtostructurenegotiationsforprotectionofecosystemservices

hasreceivedlittleattentioninliterature.CollaborativeeffortsamongprovidersandusersofSRS

are complex given the multiple objectives, mandates and perceptions characterizing involved

actors.Ourresearchquestionsrefertowhethertwo‐partynegotiationanalysiscanproveuseful

inhighlightingparties’objectivesandwhetherstructuringtheminaformaltemplatecanidentify

shareddecisionspaceswhichmightberelevantforbuildingfuturecollaboration.

Research question: Can negotiation analysis help structuring the multiple objectives

characterizingecosystemserviceprovidersandusers?Whatcharacterizesthenegotiationspace

foramechanismtoprotectSoilRegulationServicesamongprovidersandusersofSRS?

1.8Institutionalmismatchesandinformation‐sharinginmulti‐actors’andcross‐scale

governanceforSoilRegulationServices

Mandates, actions and interactions of multiple actors at different scales and from different

knowledge systems directly or indirectly constrain or promote the decisions of local actors

affecting on‐site and off‐site SRS provision (Cash andMoser, 2000). This complex multi‐actor

governance system provides potentials or barriers to improve societal response to SRS

degradationunderclimatechange.Weanalysemultipleformalpolicymandatesasdescribedby

formally stated/published objectives (e.g. in laws, decrees, etc.) to identify institutional

mismatchesandopportunities forenhancingsocietalcapacityto respond.Additionally,wetest

the use of formal governance network analysis to identify key organization that can play the

importantboundary roleneeded tobridgedifferent scalesandknowledgesystems required to

faceSRSdegradation.

25

Research questions: What are main institutional mismatches that pose barriers to collective

action in large governance network? Can formal structural analysis of governance network

identify key organizations to play boundary role in such network and thus help overcoming

institutionalmismatches?

1.9Goalsandjustificationofthemethodsused

Followingthehuman‐environmentsystemperspectiveinthefollowingparagraphs,themethods

we used helped us characterizing environmental characteristics of the provision of watershed

servicesand,forthehumansystem,approachingthedifferentdecisionprocessesatspecificscale

ofanalysis.Ihereprovidesomeadditionalrationaleforthemethodsused,sendingthereaderto

eachspecificchapterforfurtherdetailsonthemethodsused.

Firstpaperstatus:published

Locatelli B, Vignola R. 2009. “Managingwatershed services of tropical forests and plantations:

Canmeta‐analyseshelp?”.ForestEcologyandManagement,258(9),1864‐1870.

A challenge facing managers and decision makers is the complexity of the effect of forest

ecosystemsonwaterflowsalsoduetolimitedavailabilityoflarge‐nstudiesacrosssitestodetect

important trends in this relationship.Meta‐analysis is a statistical technique for combining the

quantitativefindingsofseveralstudiesandprovidesaconclusivesynthesisofthemaineffectsof

a treatment. This technique has seldom been used in the study of water‐related ecosystem

services(Ilstedtetal.,2007).

The goal of the first paper was to test the usefulness of quantitative meta‐analysis in

summarizingthefindingsofseveralstudiesthat consideredtherole forestcoverchange inthe

provisionofwatershedservices.Themainexpectedbenefitof thisresearch is toovercomethe

small“n”problemassociatedtomanyofthesetypesofstudiesgiventhehighcostofrunningthe

requiredexperiments(e.g.changingforestcover)inwatershedsespeciallyintropicalareas.

Secondpaperstatus:published

26

VignolaR.,Koellner,T.,ScholzR.W.,McDanielsT.L.,2009.Decisionmakingbyfarmersregarding

ecosystemservices:factorsaffectingsoilconservationeffortsinCostaRica.LandUsePolicy,Land

UseePolicy27,1132‐1142

The second paper addresses the complexity of farmers’ decision‐making in respect to soil

conservation decisions. We test the hypothesis on the influence of cognitive and territorial

variables(alongwithcommonly‐usedeconomicvariables)onindividualfarmers’decision‐making

regardingsoilconservation.Topreparethestructuredsurvey instrumentwedidnotpre‐define

the items and multiple response options. Instead, we built some of the items and the

correspondentmultiplechoiceoptionsthroughpreviousmeetingswhereweelicitedparticipants’

perspectivesandunderstandingonsoilerosionanditssolutions.

Thirdpaperstatus:secondroundrevisioninEcologicalEconomics

Vignola,R.,McDaniels,T.M.,Scholz,R.W.,2010.Negotiationanalysisformechanismstodeliver

ecosystemservices:avalue‐focusedbargainingstructuretoachievejointgainsintheprovisionof

SoilRegulationServices.SubmittedtoEcologicalEconomics.

Thegoalof thethirdpaperwastoanalyzewhataspectsarerelevanttoprovidersandusersof

SRS in the specific case‐study watershed. We approach this by using specific methods from

decisionscience(e.g.structuringmeans‐endsobjectives;Keeney,1992)andnegotiationanalysis

(e.g. templatebuilding;Raiffaetal., 2002). In thecontextofwatershed servicesprovidersand

usersof SRSare separatedadministrativelyandgeographicallyandmightnotknow theycould

improve their gains by collaborating. Moreover, how to structure this collaboration is not

obvious. Inthisrespect, themethodsused impliedconsultationswithstakeholderstostructure

their fundamental objectives and identify key aspects that alternatives should include. This

helpedusbuildingastakeholders’salientnegotiation‐frameworkincludingasetofalternativesto

elicittheirpreferences.

Fourthpaperstatus:submitted

Vignola, R., McDaniels, T.M., Scholz, R.W., 2010. Governance across‐scales: regulatory

mismatches and the role of boundary organizations for soil regulation‐services under climate

changeinCostaRica.SubmittedtoGlobalEnvironmentalChange.

Inthelastpaper,thegoalistoidentifyinstitutionalmismatchesarisingfromformalpoliciesand

mandates and constraining SRS provision and use. On the other side, we test how structural

27

analysis of information‐exchange network can help identify boundary organizations thatmight

helpovercometheseinstitutionalmismatches.Actorsinthegovernancenetworkhaveeachtheir

ownofficialmandateby laworconsensusandcontributedirectlyor indirectlytoSRSprovision

through their specific role (e.g. funding, science, regulation, support to farmers, etc.). To

characterize this formal policy framework it is necessary to recompile laws, text inwebpages

statingofficialmandatesandorganization’scorporatecommunicationdocuments.Toanalysethe

informalinteractionininformation‐sharingnetwork,interviewstoactorsarerequiredinorderto

letthemindicatewithwhomtheir interchange informationandelicit theirviewsonhowother

organizations in the network produce trustable information and influence relevant policy

processes.

1.10References

Aguilar,E.,Peterson,T.C.,RamırezObando,P.,Frutos,R.,Retana,J.A.,Solera,M.,Soley,J.,GonzalezGarcıa,I.,

Araujo,R.M.,RosaSantos,A.,Valle,V.E.,Brunet,M.,Aguilar,L.,Alvarez,L.,Bautista,M.,Casta˜nonp,C., Herrera, L., Ruano, E., Sinay, J.J., Sanchez, E., Hernandez Oviedo, G.I., Obed, F., Salgado, J.E.,Vazquez, J.L., Baca, M., Gutierrez, M., Centella, C., Espinosa, J., Martınez, D., Olmedo, B., OjedaEspinoza,C.E.,Nu˜nez,R.,Haylock,M.,Benavides,H.,Mayorga,R.,2005.Changesinprecipitationandtemperature extremes in Central America and northern South America, 1961–2003. Journal ofGeophysicalResearch110,D23107.

Andreassian, V., 2004. Water and forests: from an historical controversy to scientific debate. Journal ofHydrology,291:1‐27.

Bonell,M.,Possibleimpactsofclimatevariabilityandchangeontropicalforesthydrology.ClimaticChange39,215‐272

Calder, I.R.,2002.Forestsandhydrologicalservices:reconcilingpublicandscienceperceptions.LandUseandWaterResourcesResearchNo.2,pp.2.1–2.12.http://www.luwrr.com.

Cash, D.W., Clark, W.C., Alcock, F., Dickson, N.M., Eckley, N., Guston, D.H., Jager, J., Mitchell, R.B., 2003.Knowledge systems for sustainable development. Proceedings of the National Academy of Science(PNAS)100(14),8086‐8091.

Cash, D.W., 2001. In order to aid in diffusing useful and practical information: agricultural extension andboundaryorganizations.Science,TechnologyandHumanValues26(4),431‐453.

Cash,D.W.,MoserS.C.,2000.Linkingglobalandlocal scales:designingdynamicassessmentandmanagementprocesses.GlobalEnvironmentalChange10,109‐120.

Chesworth,W.,2008.EncyclopediaofSoilScience.Springer,TheNetherlands.Chomitz,K.M.,Kumari,K.,1996.CentreforSocialandEconomicResearchontheGlobalEnvironment,Working

PaperGEC96‐19,Worldbank,Washington,USA.Duit,A.,Galaz,V.,2008.Governanceandcomplexity:emergingissuesforgovernancetheory.Governance:an

internationaljournalofPolicy,AdministrationandInstitutions21(3),311‐335.Fisher, B., Turner, R.K., Morling, P., 2009. Defining and classifying ecosystem services for decision making.

EcologicalEconomics68,643‐653.

28

Gurevitch,J.,Curtis,P.S.,Jones,M.H.,2001.Meta‐analysisinEcology.AdvancesinEcologicalResearch,32:199‐247.

Hulme,P.E.,2005.Adaptingtoclimatechange:istherescopeforecologicalmanagement inthefaceofglobalthreat?JournalofAppliedEcology42,784‐794.

Ilstedt,U.,Malmer,A.,Verbeeten,E.,Murdiyarso,D.,2007.Theeffectofafforestationonwater infiltrationinthetropics:asystematicreviewandmeta‐analysis.ForestEcologyandManagement251(1–2),45–51

Innes, J.E., Booher, D.E., 1999. Consensus building and complex adaptive systems. Journal of The AmericanPlanningAssociation65(4),412‐423.

Keeney,R.L.,1992.Value‐focusedthinking:apathtocreativedecision‐making.HarvardUniversity Press,432pp.

McNeill,J.R.,Winiwarter,V.,2004.Breakingthesod:humankind,history,andsoil.Science304,1627‐1629.Morgan,R.P.C.,2005.Soilerosionandconservation.BlackwellScienceLtd,USA.304pp.Montgomery,D.R.,2007.Soil erosionandagricultural sustainability.Proceedingsof theNationalAcademyof

Science104(33),13268‐13272.Oldeman,L.R.,1992.Theglobalextentofsoildegradation.In:Greenland,D.J.,Szabolcs,I.(Eds.),SoilResilience

and Sustainable Land Use. Proc. Symposium, Budapest, Hungary. 28 Sept.‐2 Oct. 1992, CABInternational,Wallingford,UK,pp.99‐l18.

Orts, W.J., Roa‐Espinoza, A., Sojka, R.E., Glenn, G.M., Imam, S.H., Erlacher, K., Pedersen, J.S., 2007. Use ofSynthetic Polymers and Biopolymers for Soil Stabilization in Agricultural, Construction, and MilitaryApplications.JournalofMaterialsinCivilEngineering19(1),58‐66.

Pagiola,S.,PaymentsforEcosystemServicesinCostaRica,EcologicalEconomics65(4),712‐724.Pimentel,D.,Kounang,N.,1998.Ecologyofsoilerosioninecosystems.Ecosystems1,416‐426.Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., McNair, M., Crist, S., Shpritz, L., Fitton, L.,

Saffouri, R., Blair, R., 1995. Environmental and economic costs of soil erosion and conservationbenefits.Science267(5201),1117‐1123.

Raiffa,H.,Richardson,J.,Metcalfe,D.,2002.Negotiationanalysis:thescienceandartofcollaborativedecision‐making.HarvardUniversityPress,548pp.

Riedel, M.S.; Vose, J.; Bolstad, P. 2004. Characterizing Hysteretic Water Quality in Southern AppalachianStreams.InProceedingsofNationalWaterMonitoringCouncil,Chattanooga,TN,USA.

Santibaniez,F.,Santibaniez,P.,2007.Trends in landdegradationinLatinAmericaandtheCaribbean,theroleof climate change. In Sivakumar, M.V.K., Ndiangui, N. Eds. 2007. Climate and Land Degradation.Springer‐Verlagpublisher,Berlin,Germany.

Schoenholtz,S.H.2004. Impactsofforestmanagementonwaterquality. InHydrology.EnciclopediaofForestSciences.Availableonline:www.elsevier.com/locate/.doi:10.1016/B0‐12‐145160‐7/00209‐X.

Scholz, R.W., Binder, C., 2003. The paradigm of human‐environment systems. Social and Natural ScienceInterface‐ETHZ,WorkingPaper37.

Sivakumar,M.V.K.,Ndiangui,N.,2007.Climateandlanddegradation.SpringerlinkPublisher,624pp.WorldResourcesInstitute(WRI),2000.WorldResources2000‐2001:Peopleandecosystems:Thefrayingweb

oflife.WorldResourcesInstitute(WRI),WashingtonD.C.,US.UNESCO, WMO, 1992. International glossary of hydrology. Paris, United Nations Educational Scientific and

CulturalOrganization;Geneva,WorldMeteorologicalOrganization,1992.VandeBergh,J.C.J.M.,Ferrer‐i‐Carbonell,A.,Munda,G.,2000.Alternativemodelsofindividualbehaviourand

implicationsforenvironmentalpolicy.EcologicalEconomics32,43‐61.Wischmeier,W.H.,Smith,D.D.,1978.PredictingRainfallErosionLosses.AGuidetoConservationPlanning.US

DepartmentofAgricultureHandbookNo.537.USGovernmentPrintingOffice,WashingtonD.C.,USA.

30

2. Managing watershed services of tropical forests and plantations: can meta-analyses help?

Paperpublished:LocatelliB,VignolaR.2009.“Managingwatershedservicesoftropical forests

andplantations:Canmeta‐analyseshelp?”.ForestEcologyandManagement,258(9),1864‐1870.

Abstract

Thewatershed services provided by tropical natural and planted forests are critical to human

well‐being. An increasing number of valuation studies and experiences with payment for

ecosystem services have dealt with the role of ecosystems in regulating the flow of water.

However, several studies have been based on misconceptions about the role of forests and

plantationsinthehydrologicalcycle,despitethepublicationofmanyreviewsbyhydrologists.The

objectiveofthispaperistoevaluatewhethermeta‐analysesappliedtostudiescomparingwater

flows in tropicalwatershedsundernaturalorplanted forestsandnon‐forest lands canprovide

usefulresultsforvaluingwatershedecosystemservicesandmakingdecisions.Themeta‐analyses

show significantly lower total flows or base flows under planted forests than non‐forest land

uses. Meta‐analyses conducted with subsamples of the data also show lower total flow and

higherbaseflowundernaturalforeststhannon‐forestlanduses.However,theavailablestudies

were restricted to humid climates and particular forest types (Pinus and Eucalyptus planted

forestsandlowlandnaturalforests).Thesmallnumberofavailablestudieswithsufficientoriginal

dataisamajorconstraintintheapplicationofmeta‐analyses.Thisrepresentsamajortechnical

challengeforvaluationstudiesorpaymentforecosystemservices,especiallyincountrieswhere

financialresourcesforimplementingfieldresearcharescarce.

Keywords:ecosystem services; tropics; natural forest; planted forests; hydrology; policy;meta‐

analysis

2.1Introduction

Ecosystems provide services critical to humanwell‐being, in particularwatershed services that

regulate the quantity of water available for human activities. The conservation of dry season

streamflowsisessentialfornavigation,recreation,wildlife,andforruralcommunities,aswellas

for irrigation systems that lack the technology for pumping groundwater (Aylward, 2005). The

reductionofstormflowmaybenefithousing,infrastructure,oragricultureinflood‐proneareas.

31

Theconservationof total annualwater flow isalso relevant to reservoirs fordrinkingwateror

hydroelectricityproduction(Guoetal.,2000).

Over thepast50years,however, theconversionofnaturalecosystems toother landuseshas

dramatically altered hydrological cycles (Millennium Ecosystem Assessment, 2005). The

combined effects of climate and land cover changes require societies to adopt appropriate

adaptation measures for reducing their vulnerability to water scarcity and excess (Oki et al.,

2006;Hulme,2005).Thesemeasuresshouldincludetheprotectionorrestorationofecosystems

providingwatershedservices,especiallyindevelopingcountrieswithlowtechnicalandfinancial

capacityto regulatewater flowswithengineeredsolutions(Bergkampetal., 2003;Pattanayak,

2004).Inthisrespect,financialmechanismsthatencouragetheprovisionofecosystemservices,

suchaspaymentsforecosystemservices(PES),areincreasinglybeingusedtomanageupstream

forest ecosystems for the regulation of water flows (Wunder, 2005; FAO, 2004; Dudley et al.,

2003).

Valuation studies ofwatershed ecosystem services andmanagement or policy decisions about

PESarenotalwaysscientificallysound.Variousmisconceptionsabouttheroleofecosystemsin

regulating the flow of water persist among managers and decision‐makers, despite the

publication of many scientific papers on this issue (e.g., Bosch and Hewlett, 1982; Bruijnzeel,

1990;CritchleyandBruijnzeel,1996;SahinandHall,1996;Bonell,1998;Calder,2002;Bestetal.,

2003;Andreassian,2004;Bruijnzeel,2004;Scottetal.,2004;BonellandBruijnzeel,2005;Farley

et al., 2005; Guillemette et al., 2005; Scott et al., 2005). A challenge facing managers and

decision‐makers is thecomplexityoftheeffectofecosystemsonwater flows(Bruijnzeel,1990;

Fujiedaetal.1997;Bonell,1998;vanNoordwijketal.,2004;Waageetal.,2008).).

Meta‐analysis,astatistical techniqueforcombiningthequantitativefindingsofseveralstudies,

hasseldombeenused in foresthydrology(Ilstedtetal.,2007).Meta‐analysishastheabilityto

considerseveralstudiesthatareinthemselvesinconclusive,andprovideastatisticallyconclusive

synthesis.Althoughitrequiressimplificationoftheobservedphenomena,meta‐analysishasthe

advantage of producing results that are more easily understandable by decision‐makers than

narrative reviews. The objective of this paper is to evaluatewhethermeta‐analyses applied to

studiescomparingwaterflowsintropicalwatershedsunderdifferentforestornon‐forestcovers

32

canprovideusefulresultsforvaluingwatershedecosystemservicesandmakingdecisionsabout

PESandwatershedmanagement.

2.2Method

Weconductedseveralmeta‐analysestocombinethefindingsofstudiescomparingwaterflows

betweenwatershedsundernaturalforestsvs.non‐forestlanduses,andplantedforestsvs.non‐

forestlanduses.Thesestudiesweresynchroniccomparisonsoftwoormorepairedwatersheds

or diachronic comparisons of one watershed under changing forest cover over time. We

considerednaturalandplantedforestsbecausebotharebeingconsideredinPESforwatersheds

(Wunder et al., 2008). For instance, the national PES scheme in Costa Rica rewards forest

conservation and plantation and recognizes explicitly hydrological services, including the

provisionofwaterforhumanconsumption,irrigation,andenergyproduction(Pagiola,2008).

Weconsideredthreehydrologicalvariablesof interest:annualtotal flow,stormflow,andbase

flow.According to theGlossaryofHydrologyofUNESCO‐WMO(1992), annual total flow is the

“total volumeofwater that flowsduringayear,usually referring to the outflowofadrainage

area or river basin”. Storm flow is “part of surface runoffwhich reaches the catchment outlet

shortlyaftertherainstarts; itsvolume isequal torainfallexcess”.Baseflow is the“partofthe

discharge(volumeofwaterflowingthroughariverorchannelcross‐sectioninunittime)which

entersastreamchannelfromgroundwaterfromlakesandglaciersduringlongperiodswhenno

precipitation or snowmelt occurs” (UNESCO‐WMO, 1992). Other authors have modified this

definition by adding that base flow is “themore or less permanent flow supplied to drainage

channelsbyratherinvariablesources”(Sussweinetal.,2001).Inareaswithseasonalityinrainfall,

thedischargeduringmostofthedryseasonresultsfrombaseflow(Smakhtin,2001).

We followed the generally accepted procedures for meta‐analysis (e.g., Cooper and Hedges,

1994;Gurevitchetal., 2001). First,we searchedpeer‐reviewedarticles comparingwater flows

undernaturalorplanted forestsandnon‐forest landuses in tropicalwatersheds.Wesearched

referencesinCAB,ISIWebofScience,andtheAGRICOLAliteraturedatabaseinMarch2006with

thefollowingquery:“(forestORdeforestationORreforestationORafforestation)AND(waterOR

hydrology OR hydrological) AND (watershed OR catchment OR land use OR land‐use) AND

(tropicalORsubtropical)”.About1100referenceswereretrieved.

33

Second,we selected studies conducted inwholewatersheds (i.e., not in plots) that compared

several watersheds under different land‐uses during several years or one catchment during

several years before and after a land‐use change. We selected only studies reporting field

measurements(i.e.,notmodelingresults)andproviding“sufficientdata”.By“sufficientdata”,we

mean that thestudies reportedoneof the followingdata combinations: (1)annual valuesofa

hydrologicalvariableforeachwatershedandeachyear,or(2)meansandstandarddeviationsof

ahydrological variableand thenumberofobservations forwatershedsunder forestsandnon‐

forest land uses. This selection resulted in only 10 studies. To ensure a larger sample, we

searched for documents cited by review articles that compliedwith our selection criteria, and

retrieved 10more studies.We characterized the studies according to forest type (planted vs.

natural forests), watershed area, humidity index, and forest cover difference. Data about the

humidity index, defined as the ratio of annual precipitation and potential evapotranspiration,

camefromDeichmannandEklundh(1991).Forestcoverdifferencewasdefinedasthedifference

inforestcoverbetweenthecomparedwatersheds,andwasconsideredtobehighifitwasmore

than50%ofthewatershedarea.

Third, each comparison (i.e. comparison of one hydrological variable between different

watersheds) found in the selected studies was converted to a dimensionless scale called the

effect size. We chose the Hedge's unbiased estimator g of effect size, and used equations

described by Cooper and Hedges (1994) to estimate the effect size and its variance for each

comparison.Fourth,wecombinedtheestimatesofeffectsizesusingarandomeffectmodelthat

considers a random variation among the studies in the "true" effect (Gurevitch et al., 2001;

ShadishandHoaddock,1994).Weconductedsixmeta‐analysesforeachofthethreehydrological

variables of interest and for natural and planted forests. We also applied meta‐analyses to

subsetsofthedata,e.g.,onlysmallwatersheds.Finally,foreachmeta‐analysis,weconducteda

sensitivityanalysistodeterminewhethersomeindividualcomparisonsinfluencedtheresults,by

performingn partialmeta‐analyses (n being the number of comparisons) on data sub‐samples

containingallthecomparisonsbutthenthone.Onlysignificantdifferences(atp<0.05)observed

inameta‐analysisandall associatedpartialmeta‐analyseswere reported.We considered that

thetendencytoreportonlysignificantresults–orthe"filedrawerproblem"(Fernandez‐Duque

34

andValeggia,1994) ‐wasnotarelevantbias inouranalysis,becausetheresultsofcostlylong‐

termhydrologicalstudiesaregenerallyreportedevenifnosignificantdifferencesarefound.

2.3Resultsanddiscussion

2.3.1Selectedstudies

Amongthe20selectedstudies,ninewereconducted inAsia,eight inAfrica,andthree inLatin

America(seeTable1fordetailsofthestudies,andFigure1fortheirlocation).Thesmallnumber

of studies is due to the lack of comparison betweenwatersheds under forests and non‐forest

land uses and the lack of sufficient data in many studies. The focus of the meta‐analyses on

tropicalareas, thethemeofthisspecial issueofForestEcologyandManagement,alsostrongly

reducedthenumberofavailablestudies,astropicalareasareunder‐representedinthescientific

literatureaboutforesthydrology.Forexample,among135watershedexperimentsreviewed in

Andreassian(2004),only16aretropicalwatersheds.

35

Table 1. Characteristics of the selected studies.

Firstauthorandyear

CountryHumidityclass(2)

Watershedarea(3)

Forestcoverdifference(4)

Years(5) Description(6)

Bailly1974p(1) Mada‐gascar

HU S Hi 64‐71

NFSyn(4w.forest,2w.agriculture/fallow)

PFSyn(1w.Eucalyptus,2w.agriculture/fallow)

Bailly1974m Mada‐gascar

HU S Hi 62‐72PFSyn(1w.Eucalyptus,2w.

grassland)Bewket2005 Ethiopia

HU L Lo60‐64and80‐84

NFDia(1w.lessforestedin1960thanin1982).

Blackie1972 KenyaHU‐DR L Hi 61‐68

NFSyn(1w.forest,1w.partiallyagriculture)

Chandler1998 ThePhili‐ppines

HU S Hi 95‐96NFSyn(1w.forest,2w.

grasslands)Costa2003 Brazil

HU L Lo49‐68and79‐98

NFDia(1w.lessforestedin49‐68thanin79‐98).

Dagg1965 TanzaniaHU S Hi 57‐63

NFSyn(1w.forest,1w.agriculture)

Fritsch1983 FrenchGuyana

HU S Hi 77‐81NFSyn(2w.forest,6w.deforested/pasture)

Fritsch1992 FrenchGuyana

HU S Hi 77‐87

NFSyn(2w.forest,2w.pasture/slash‐and‐burn)

PFDia(2w.beforeandafterreforestation)

Goujon1968 Mada‐gascar

HU S Hi 62‐66 PFSyn(1w.Pinus,1w.grassland)

Lal1997 NigeriaHU‐DR S Hi

74‐75and79‐84

NFDia(1w.deforestedin1979)

Mathur1976 India

DR S Hi61‐67and69‐73

PFSyn(after19691w.Eucalyptus,1w.shrub)

PFDia(1w.undershrubbefore1968,thenEucalyptus)

Mungai2004 SriLankaHU L Lo

51‐61,67‐77,78‐88

NFDia(1w.underdeforestation)

Mwendera1994 MalawiHU L Hi

61‐65,70‐78

PFDia(1w.underforestationwithPinusandEucalyptus)

Raghunath1970 IndiaHU L Hi

Variousperiods

NFSyn(9w.mainlyunderagricultureorpasture,3w.

mainlyunderforest)Samraj1988;Sharda1998;Sikka2003

India

HU S Hi68‐71,73‐81,82‐91

PFSyn(after19721w.Eucalyptus,1w.shrub)

PFDia(1w.undershrubbefore1972,thenEucalyptus)

Wilk2001 ThailandHU L Hi

57‐64,87‐94

NFDia(1w.withdecreasingforestcover)

Zhou2002 ChinaHU S Hi 81‐90

PFSyn(1w.Eucalyptus,1w.baresoil)

36

(i) (1)Twostudiesarereportedinthesamereference(Baillyetal.,1974),oneinPérinet(1974p)andanotherinManankazo(1974m).

(ii) (2)HU=humidarea,HU‐DR=humidareanear the transitionbetweenhumidanddry (less than50km),DR=dryarea.Humidity indextakenfromDeichmannandEklundh(1991).

(iii) (3)S=Small(<1km2),L=Large(>1km2).

(iv) (4)Lo:forestcoverdiffersbylessthan50%ofthewatershedareabetweenthecomparedwatersheds,Hi:morethan50%.

(v) (5)Period(s)oftimeselectedforouranalysis.(vi) (6)NF:comparisonsbetweennatural forestsandnon‐forest landuses,PF:comparisonbetweenplanted forestsandnon‐forest landuses.Syn:synchroniccomparisons.Dia:diachroniccomparisons.Inparenthesis:descriptionofwatershedlandcover(“w.”=watershed).

Amongthe20studies,13studieswereconductedinwatershedssmallerthan1km2andsevenin

watersheds larger than 1 km2. Seventeen studies comparedwatershedswithhigh forest cover

difference(morethan50%ofthewatershedarea).Seventeenstudieswereconductedinhumid

areas,andthreeindryareasortransitionareasbetweenhumidanddry.Thisresultshowsabias

in the meta‐analyses, as no studies in semi‐arid or arid areas and few studies in dry areas

provided sufficient data. As the results cannot be generalized to the tropics as a whole, the

scientificknowledgefordecisionmakingindry,semi‐arid,andaridareasislacking.

Figure 1. Location of the 20 studies selected for the meta-analyses.

In the 20 studies,we retrieved 63 comparisons betweenwatersheds under natural or planted

forestsandnon‐forestlanduses.Onaverage,acomparisonwasbasedon16observations(e.g.,

onewatershedobservedduring16yearsorfourwatershedsduringfouryears).Thenumbersof

studies and comparisons arewithin the acceptable range of sample sizes used in othermeta‐

analyses.Indeed,inthefewmeta‐analysesthathavebeenappliedtohydrology,thesamplesizes

weregenerallylower;forinstance,14observationsandfourstudiesinIlstedtetal.(2007).

37

2.3.2Naturalforests

In 12 of the 20 selected studies, 39 comparisons were analyzed between watersheds under

naturalforestsandnon‐forestlanduses.Thesecomparisonswererelatedmostlytowatersheds

largerthan1km2(26comparisons),withhighforestcoverdifferences(24),and inhumidareas

(36). No studies about montane cloud forests complied with our selection criteria. As these

forestscanhaveadifferenteffectonwaterflowscomparedtoothernaturalforestsbecauseof

their capacity to collect water from the clouds, valuation of ecosystem services or decision

makingaboutcloudforestsshouldnotbebasedonourmeta‐analyses.

Including all comparisons, the meta‐analysis did not show significant differences in total flow

between watersheds under natural forests and non‐forest land uses (see Table 2). However,

whenselecting comparisonsmade in smallwatershedsor largedifferences in forest cover, the

meta‐analyses showed that the total flowwas significantly lower inwatersheds under natural

foreststhannon‐forestlanduses.Thisresultshowstheinterestincomparingsmallwatersheds

orwatershedswithlargedifferencesinforestcover.Someauthorshavestatedthattheeffectsof

forestsonwaterflowsarediscernableonlyinwatershedssmallerthan1000km2(FAOandCIFOR,

2005) or 500 km2 (Pattanayak, 2004), while others have stated that the most significant

observationshavebeenmadeinsmallwatersheds,usuallylessthanonekm2(Bruijnzeel,2004).

Separating the hydrological effects of forests is more difficult in larger catchments, which

generallypresentdiverselandusesandland‐usechanges.

38

Table 2. Results of the meta-analyses. Results of the meta‐analysiswithalldata

Selectedstudies(i) Number of comparisonsandcharacteristicsofthecomparedwatersheds(ii)

Othersignificantresultsofthemeta‐analyses with datasubsets(ii)

Differencesinflowsbetweennaturalforestsvs.non‐forestlanduses(Studiesavailableonlyforlowlandforests.Nostudiesoncloudforests).

Totalflow:Nosignificantdifference

11(Bai74pBew05Bla72Cha98Cos03Dag65Fri83Lal97Mun04Rag70Wil01)

15(6smallw,9largew,1 dry,12humid,2 11high3 diff.,4lowdiff)

Withsmallw.orhighdiff.:Lesstotalflowinnaturalforestthannon‐forestlanduses

Baseflow:Nosignificantdifference

4(Bai74pBew05Mun04Wil01)

8(1smallw,7largew,8humid,2highdiff.,6lowdiff.)

Inlargew.:Morebaseflowinnaturalforestthannon‐forestlanduses

Stormflow:Nosignificantdifference

8(Bai74pBew05Cha98Cos03Fri92Mun04Rag70Wil01)

16(6smallw,10largew,16humid,9highdiff.,7lowdiff.)

None

Differencesinflowsbetweenplantedforestsvs.non‐forestcover(StudiesavailableonlyaboutEucalyptusandPinusplantations.

Nostudiesonotherspecies,incl.nativespecies).Totalflow:Lesstotalflowinplantedforestthannon‐forestlanduses

7(Bai74mBai74pGou68Mat76Sam88Sha98Zho02)

11(11smallw,1dry,10humid,11highdiff.)

Withsmallw.,humidorhighdiff.:Sameresult

Baseflow:Lessbaseflowinplantedforestthannon‐forestlanduses

5(Bai74pMwe94Sam88Sha98Sik03)

8(7smallw,1largew,8humid,8highforestdiff.)

Withsmallw.,humidorhighdiff.:Sameresult

Stormflow:Nosignificantdifference

5(Bai74pGou68Mwe94Sam88Sha98)

4 (4smallw,1largew,5 5humid,5highdiff.)7

None

(vii) Numberofstudiesandreferences inparenthesis(referencesaregivenwiththe first threelettersof thefirstauthor’snameandthe2‐digityear).

(viii) “smallw.”=watershedsmallerthan1km2,“largew.”=watershedlargerthan1km2,“humid”=humidaccordingtoTable1definition,“dry” =dryor transition fromhumid todry, “low diff.”= forest coverdiffers by less than50%of thewatershedarea between the comparedwatersheds,“highdiff.”=morethan50%.

The effect of natural forests on total flowhas been demonstrated by other authors in diverse

situations, although theymay not be valid for some very old forests or cloud forests (Calder,

2002;Bruijnzeeletal., 2004).Theseeffects can beexplainedby thehigherevapotranspiration

ratesinforestscomparedtopastureorannualcroppinglanduses(Bruijnzeel,1990).Treeswith

deeprootsandhightranspirationratesmayactaspumpsthatremovewaterfromthesoiland

transpire it. Although conversion of natural forests to non‐forest land uses is immediately

followedbyaperiodofincreasedtotalflow,thesubsequentperiodmaynotbecharacterizedby

a high total flow, for instance with forest recovery or high evapotranspiration rates in the

subsequentvegetation.

39

The meta‐analysis showed no significant differences in base flow between watersheds under

naturalforestsandnon‐forestlanduses(seeTable2).Theeffectofnaturalforestsonbaseflow

resultsfromtwocompetingprocesses(Calder,2002;Bruijnzeeletal.,2004):hightranspirationby

the forest, contributing to a low base flow, and high infiltration under the forest, generally

contributing to soilwater rechargeandhighbase flow. If the infiltrationcapacityof the soil is

increasedwhenconvertingnatural foreststonon‐forest landuses,baseflowcanbe increased.

Natural forests conserve base flow compared to other land uses in situations where the

alternative land use decreases the infiltration capacity (Bruijnzeel et al., 2004). In the meta‐

analysis,thebaseflowwashigherundernatural forestsforthesubsetoflargewatersheds;but

this result is difficult to explain. This may due to the regional recycling of rainfall by forests

(Vanclay,2009)butthisinterpretationiscontroversial(Bruijnzeel,2004).

Regarding storm flow, no significant difference was found betweenwatersheds under natural

forests and non‐forest land uses (see Table 2). Forest hydrologists agree that the relationship

betweenforestsandstormflowisnotclear(BoschandHewlett,1982).Evenifhigherinfiltration

andevapotranspirationunderforestsmayreducestormflow,ourresultsshowedthatthiseffect

wasnotalwayssignificant.Theroleofforestcoverforstormflowreduction isadebatedissue,

especiallyforthepartofthestormflowthatisreleasedduringextremerainfallevents(peakflow

ormaximumflow)andisresponsibleforfloods.AccordingtoFAOandCIFOR(2005),thereisno

hardevidencethattropicalforestsreducefloods,whileBradshawetal.(2007)showedthatflood

frequency isnegatively correlatedwith forest cover.Theeffectof forestson large scale floods

seemsparticularlyweak;Bruijnzeel(2004)concludedthatfloodsoccurringinextremeconditions

(e.g., extreme rainfall at theendof the rainy seasonwhensoilsarewet)arenot regulatedby

vegetationcover.Ouranalysesdidnotdistinguishstormflowsfromrainfalleventswithdifferent

intensities,andthus itwasnotpossibletodeterminethe roleof forests inregulatingthepeak

flowresultingfromextremerainfallevents.Theregulationofannualstormflowisdifferentfrom

the regulation of peak flow resulting in large‐scale floods (Bruijnzeel, 2004). However, even if

forests do not prevent large scale floods, their role in preventing average andmost frequent

floodsshouldnotbeundervalued.

40

2.3.3Plantedforests

In10ofthe20selectedstudies,24comparisonsbetweenwatershedsunderplantedforestsand

non‐forest land uses were analyzed. These comparisons were related mostly to watersheds

smaller than 1 km2 (22 comparisons),with high forest cover differences (all 24), and in humid

areas (23). The selected studies referred to planted forests with species of two genera:

EucalyptusandPinus.Nostudieswereavailableforotherspecies,includingnativespecies.This

showsthatthereisalackofscientificdataformakingdecisionsaboutplantingspeciesotherthan

EucalyptusandPinusgenera,as the resultsof themeta‐analyses cannotbegeneralized toany

plantedforestinthetropics.

Themeta‐analysesshowedthattotalflowandbaseflowwerelowerinwatershedsunderplanted

forests than non‐forest land uses (see Table 2). These results can be explained by the high

transpiration rates of exotic species, especially Eucalyptus (Vertessy et al., 2001). Infiltration

capacitymaybehigherunderplantedforeststhannon‐forestlanduses(Ilstedtetal.,2007),but

maynotbeenoughtooffsetthehigher lossofwaterbytranspiration.However,someauthors

have reported that in degraded soils, planted forests may increase infiltration more than

transpiration, thus increasing base flow, compared to non‐forest land uses (Bruijnzeel, 2004).

However, theavailablestudiesdidnotcomparewater flowsbetweenplantedforestsandnon‐

forest land uses on degraded soils. Including all comparisons, themeta‐analysis did not show

significant differences in storm flow betweenwatershedswith planted forests and non‐forest

landuses.Thisanalysisofplantedforestsandstormflowwasbasedononly fivecomparisons,

comparedtoeightto16forotheranalyses.

2.3.4Usefulnessofthemeta‐analyses

Theresultsofthemeta‐analysesallowtheanalysisofsomemisunderstandingsabouttheeffects

of natural and planted forests on water flows in decision making on PES or watershed

management. According to the conventional beliefs analyzed by various authors (Chomitz and

Kumari, 1996; Susswein et al., 2001; Calder, 2002; Bruijnzeel et al., 2004; Kaimowitz, 2005),

natural and planted forests are thought to increase total flows. This belief is deeply rooted in

publicperception,aswasshownbyKosoyetal.(2007)inCentralAmerica,wheremorethan90%

41

of the population involved in a survey perceived thatmore forestswould lead to higher total

waterflows.AccordingtoFAO(2004),manyhydrologicalPESarebasedonassumptionsthathave

notbeenverifiedintheparticularcase.Forexample,inthePESschemeinPimampiro(Ecuador),

theeffectofland‐useofhydrologicalserviceshasnotbeenstudied(WunderandAlbán,2008).In

CostaRica,aprivatehydro‐energyfirmpaysforecosystemservicesprovidedbyforestationand

forest conservation with the objective of increasing its energy production, in part through

increasingtotalflow(Mirandaetal.,2007).ThreePESschemesstudiedbyKosoyetal.(2007)in

CentralAmericaarebuiltuponthebeliefthatforestsincreasetotalflows.

Aylward(2005)showedthatseveralvaluationsofforesthydrologicalserviceswerebasedonthe

unverified assumption that natural forestswould decrease storm flow and increase base flow.

AmongthestudiesvaluingforestwatershedservicesintheFAOForestValuationDatabase(FAO,

2007), three were based on unsubstantiated relationships between forests and water, for

instance that forests decrease floods or increase total flows. To supply water and reduce

sedimentinthePanamaCanal,aplandevelopedinthe1990spromotedplantingforestsamong

otheractivities,andwasbasedonthebeliefthatplantedforestswouldincreasetotalandbase

flows(Calder,2002;Kaimowitz,2005).AfterhurricaneMitchinCentralAmericainOctober1998,

thepublic,decision‐makers,environmentalists, and internationalagencieswereconvinced that

deforestation had increased the damage, and proposed forest planting as a watershed

managementmeasure (Kaimowitz, 2005). Similarly, Bruijnzeel (2004) reported that large‐scale

forestplanting inupperwatershedswasproposed in BangladeshandChina to reducedamage

causedbyfloods.

Even though the results of the meta‐analyses should not be generalized to any tropical

watershed,totalflowsappearedtobelowerinwatershedsundernaturalforeststhannon‐forest

land uses (for smallwatersheds orwith high differences in forest cover). The results showno

significantdifferences in storm flowbetweenwatershedsunder natural forestsandnon‐forest

landuses.Regardingplantedforests, theresultsshowedthattheyreducedbaseflowandtotal

flowandhadnosignificanteffectonstormflowcomparedtonon‐forestlanduses.Theseresults

highlight that the effects of natural or planted forests on water flows are not the same as

perceivedbysomedecisionmakers.Meta‐analysescanhelpcorrectmisunderstandingsandshow

42

that some of the effects of natural or planted forests on water flows are unclear, and that

inappropriategeneralizationsshouldbeavoided(Tognettietal.,2004).

2.3.5Limitations

Themeta‐analysesfacedseveral limitations.First, thesmallnumberofavailablestudies limited

thenumberofsignificantdifferencesandpreventedfromtaking intoaccountother factors, for

instancesoilcharacteristics,geology,topography,rainfallpattern,locationofland‐usechangein

thewatershed and landmanagement in non‐forest land areas,which are important factors in

explaining the difference in water flows between forested and non‐forested watersheds

(Bruijnzeel, 2004; Aylward, 2005). Applyingmeta‐analyses in a rigorousway is difficult with a

smallnumberofstudies,especiallywithregardstothesensitivityanalysiswhichisconductedon

subsamplesofthedataset.Thefollowingresultswerediscardedbythesensitivityanalysis:more

base flow in watersheds under natural forests than under non‐forest land uses (with all

watersheds),lessstormflowinwatershedsundernaturalforeststhanundernon‐forestlanduses

(onlywithwatershedswhereforestcoverdifferedbymorethan50%ofthewatershedarea),and

lessstormflow inwatershedsundernatural foreststhanundernon‐forest landuses(onlywith

smallwatersheds).

Second, the available studies had beenmostly conducted in humid areas and in some forest

types (e.g., no cloud forests or planted forests with species other than of the Pinus and

Eucalyptus genera). For these reasons, the results cannot be generalized to the entire tropics.

Third,theavailablestudiesuseddifferentmethods(e.g.forseparatingbaseflowandstormflow)

anddifferenttimeperiodsforcalculatingflow‐rainfallratios.Amorethoroughanalysisoforiginal

hydrologicaldatawouldbenecessarytoextendoursimplifiedmeta‐analyses.

2.4Conclusions

Meta‐analysisappearstobeapromisingwaytocombineresults fromstudiescomparingwater

flowsbetweentropicalwatershedsundernaturalorplantedforestsandnon‐forestlanduses.It

canhelpdecisionmakersunderstandtheeffectsofforestsonwaterflows,andalsohighlightthe

effects that remain unclear. This help is necessary because even though hydrologists have

published many narrative reviews about these effects, not all economists or decision makers

have attempted to integrate this knowledge into the process of determining the value of

43

hydrological ecosystem services or formaking decisions. Themeta‐analyses show significantly

lowertotal flowsorbaseflowsunderplanted foreststhannon‐forest landuses.Meta‐analyses

conductedwithsubsamplesofthedataalsoshowlowertotal flowandhigherbaseflowunder

natural forests than non‐forest land uses. However, the available studies were restricted to

humid climates and particular forest types (Pinus and Eucalyptus planted forests and lowland

naturalforests).

The small number of available studieswith sufficient data is amajor constraint in the use of

meta‐analysis foranalyzingtheeffectsofnaturalorplantedforestsonwater flows. It impedes

analyzing the interacting effect of important factors for water flow regulation, such as soil,

geology, topography, or land management practices. Furthermore, the available studies with

sufficientdataare restricted to some local conditionsor forest typesandknowledge is lacking

aboutdryareas,cloudforests,nativespeciesplantations,andplantationsondegradedlands.

The lack of measurements of water flow under different land‐uses and of empirical data on

hydrologicalservicesrepresentsamajortechnicalchallengeforvaluationstudiesorpaymentfor

ecosystemservices(KremenandOstfeld,2005),especiallyincountrieswherefinancialresources

for implementing field research are scarce. More empirical data on underrepresented local

conditions or forest types and a facilitated access to hydrological data would be valuable for

watershedmanagersanddecision‐makers.Analyzingoriginaldataaboutwaterflows,climateand

watershedcharacteristicsinaconsistentwaycouldimprovefurtherapplicationofmeta‐analysis.

2.5References

Andreassian, V., 2004. Water and forests: from an historical controversy to scientific debate. Journal ofHydrology,291:1‐27.

Aylward, B.,2005.Landuse,hydrological functionandeconomic valuation. In:M. BonellandL.A.Bruijnzeel(eds.),Forests,WaterandPeople intheHumidTropics:Past,presentandfuturehydrologicalresearchforintegratedlandandwatermanagement.CambridgeUniversityPress,pp.99‐120.

Bailly, C., Benoit de Coignac, G., Malvos, C., Ningre, J.M. and Sarrailh, J.M., 1974. Etude de 1'influence duconvert naturel et de ses modifications à Madagascar. Cahiers Scientifiques du Centre TechniqueForestierTropicalNo.4,Nogent,France,114pp.

Bergkamp,G.,Orlando,B.Burton,I.,2003.AdaptationofWaterManagementtoClimateChange.IUCN,Gland,Switzerland.

BestA,ZhangL,McMahonT,WesternA,VertessyR.2003.Acriticalreviewofpairedcatchmentstudieswithreferencetoseasonalflowsandclimatevariability.Murray‐DarlingBasinCommission,Canberra,Au.

Bewket,W.,Sterk,G.,2005.Dynamics in landcoveranditseffectonstreamflowintheChemogawatershed,BlueNilebasin,Ethiopia.HydrologicalProcesses,19:445–458.

44

Blackie, I.R.,1972.HydrologicaleffectsofachangeinlandusefromrainforesttoteaplantationinKenya.In:Proceedings of Symposium on the Results of Research on Representative and Experimental Basins,Wellington,NZ,IAHSPublicationNo.97,pp.312‐329.

Bonell, M., Bruijnzeel, L.A.,. 2005. Forests, water and people in the humid tropics: past, present and futurehydrological research for integrated land and water management. International Hydrology Series,CambridgeUniversityPress.

Bonell, M., 1998. Possible effects of climate variability and change on tropical forest hydrology. Climaticchange,39:215‐272.

Bosch, J.M, Hewlett, J.D. 1982. A review of catchment experiments to determine the effect of vegetationchangesonwateryieldandevapotranspiration.JournalofHydrology,55:3‐23.

Bradshaw,C.J.A.,Sodhi,N.S.,Peh,K.S.H.,Brook,B.W.,2007.Globalevidencethatdeforestationamplifiesfloodriskandseverityinthedevelopingworld.GlobalChangeBiology,13:1‐17.

Bruijnzeel, L.A., 1990. Hydrology of moist tropical forests and effects of conversion: a state of knowledgereview.UNESCO,Paris,230pp.

Bruijnzeel,L.A.,2004.Hydrologicalfunctionsoftropicalforests:notseeingthesoilforthetrees?.Agriculture,EcosystemsandtheEnvironment,104:185‐228.

Bruijnzeel,L.A.;Calder,I.R.;Vertessy,R.A.,2004.Foresthydrology:impactsofforestconversiononstreamflow.In:J.Burley,J.EvansandJ.A.Youngquist(eds.),TheEncyclopediaofForestSciences.Oxford,Elsevier,pp.350‐358.

Calder, I.R.,2002.Forestsandhydrologicalservices:reconcilingpublicandscienceperceptions.LandUseandWaterResourcesResearch,2:1‐12.

Chandler,D.G.,Walter,M.F.,1998.Runoffresponsesamongcommon landuses intheuplandsofMatalom,Leyte,Philippines.TransactionsoftheASAE,41(6):1635‐1641.

Chomitz,K.,Kumari,K.,1996.Thedomesticbenefitsoftropicalforests.PolicyResearchWorkingPaper1601,WorldBank,Washington,D.C.

Cooper,H.,Hedges,L.V.(eds.),1994.TheHandbookofResearchSynthesis.RussellSageFoundation,NewYork,573pp.

Costa,M.H., Botta, A.,Cardille, J.,2003. Effectsof large‐scalechanges in land coveron thedischargeof theTocantinsRiver,SoutheasternAmazonia.JournalofHydrology,283(1):206‐217.

Critchley, W.R.S., Bruijnzeel, L.A., 1996. Hydrological impacts of conversion of moist tropical forests toagricultureandplantations.IHPHumidTropicsProgrammeseriesnº10,UNESCO,Paris,50pp.

Dagg,M., Blackie, J.R., 1965. Studies of the effects of changes in land use on the hydrological cycle in EastAfricabymeansofexperimentalcatchmentareas.BulletinoftheInternationalAssociationofScientificHydrology,10(4):63‐75.

DeichmannU., Eklundh L., 1991. Global digital data sets for landdegradation studies: a GIS approach.GRIDCaseStudySeriesNo.4;UNEP/GEMSandGRID;Nairobi,Kenya;103p.(dataavailableat:http://www‐cger.nies.go.jp/grid‐e/gridtxt/hind_geo.html).

DudleyN.,StoltonS.,2003.Runningpure:theimportanceofforestprotectedareasfordrinkingwater.WorldBank,WashingtonD.C.,114p.

FAO, 2004. Payment schemes for environmental services in watersheds: regional forum, 9‐12 June 2003,Arequipa,Peru.LandandWaterDiscussionPapernº3,FAO,Roma,88pp.

FAO,2007.Forestvaluationdatabase.DevelopedundertheEC‐FAOPartnershipProgrammeonDataCollectionandAnalysis.FAO,Romaaccessedon19April2007www.fao.org/forestry/valuation

FAO,CIFOR,2005.Forestsandfloods:Drowninginfictionorthrivingonfacts?ForestPerspectivesSeriesnº2,CIFOR,Bogor,Indonesia,40pp.

Farley, KA, Jobbagy, EG, Jackson, RB., 2005. Effects of afforestation on water yield: a global synthesis withimplicationsforpolicy.GlobalChangeBiology,11:1565‐1576.

Fernandez‐Duque,E.,Valeggia,C.,1994.Meta‐analysis:avaluabletoolinconservationresearch.ConservationBiology,8(2):555‐561.

45

Fritsch, J.M., 1983. Modifications des écoulements après défrichement et aménagements agro‐forestiers debassins versants expérimentaux en Guyane française. In: Le Projet ECEREX (Guyane) : analyse del'écosystème forestier tropical humide et des modifications apportées par l'homme. Journées deCayenne,Cayenne(GF),ORSTOM,Paris,p.260‐304.

Fritsch,J.M.,1992.Leseffetsdudéfrichementdelaforêtamazonienneetdelamiseenculturesurl’hydrologiedespetitsvassinsversants.ORSTOMéditions,Paris,392pp.

Fujieda, M.; Kudoh, T.; deCicco, V.; Calvarcho, J.L. 1997. Hydrological processes at two subtropical forestcatchments:theSerradoMar,SaoPaulo,Brazil.JournalofHydrology,196:26‐46.

Goujon, P., Bailly, C., de Vergnette, J., Roche, P., 1968. Essai de mise en valeur agricole par aménagementrationneldesbassinsversants: lepérimètredeManankazosur leshautsplateauxmalgaches.BoisetForêtsdesTropiques,121:19‐34.

Guillemette, F, Plamondon, AP, Prevost, M, Levesque, D., 2005. Rainfall generated stormflow response toclearcutting a boreal forest : peakflow comparison with 50 world‐wide basin studies. Journal ofHydrology,302:137‐153.

Guo, Z., Xiao, X., Li, D., 2000.An assessment of ecosystem services: water flow regulation andhydroelectricpowerproduction.EcologicalApplications,10(3):925‐936.

Gurevitch,J.,Curtis,P.S.,Jones,M.H.,2001.Meta‐analysisinEcology.AdvancesinEcologicalResearch,32:199‐247.

HulmeP.2005.Adaptingtoclimatechange: istherescopeforecologicalmanagement inthefaceofaglobalthreat?JournalofAppliedEcology,42:784‐794.

Ilstedt,U.,Malmer,A.,Verbeeten,E.,Murdiyarso,D.,2007.Theeffectofafforestationonwater infiltrationinthetropics:Asystematicreviewandmeta‐analysis.ForestEcologyandManagement,251(1‐2):45‐51.

Kaimowitz,D.,2005.UsefulMythsandIntractableTruths:thePoliticsoftheLinkBetweenForestsandWaterinCentral America. In: M. Bonell and L. A. Bruijnzeel (eds.), Forests,Water and People in the HumidTropics: Past, present and future hydrological research for integrated land andwatermanagement.CambridgeUniversityPress,pp.86‐98.

Kosoy,N.,Martinez‐Tuna,M.,Muradian, R.,Martinez‐Alier, J.,2007. Payments for environmental services inwatersheds:InsightsfromacomparativestudyofthreecasesinCentralAmerica.EcologicalEconomics,61(2‐3),446‐455.

Kremen,C.,Ostfeld,R.S.,2005.A call to ecologists:measuring,analyzing,andmanaging ecosystemservices.FrontEcolEnviron,3(10):540‐548.

Lal,R.,1997.DeforestationeffectsonsoildegradationandrehabilitationinwesternNigeria.IV.Hydrologyandwaterquality.LandDegradationandDevelopment,8(2):95‐126.

Mathur,H.N., Babu, R., Joshie, P.andSingh, B., 1976.EffectofClearfellingandReforestationonRunoffandPeakRatesinSmallWatersheds.IndianForester,102(4):219‐26

Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well‐being: Synthesis. Island Press,Washington,DC.,137pp.

Miranda,M.,Dieperink, C.,Glasbergen, P.,2007.Voluntaryagreements inwatershedprotection experiencesfromCostaRica.Environment,DevelopmentandSustainability,9(1):1‐19.

Mungai, D.N., Ong, C.K., Kiteme, B., Elkaduwa, W., Sakthivadivel, R., 2004. Lessons from two long‐termhydrologicalstudiesinKenyaandSriLanka.Agriculture,EcosystemsandEnvironment,104(1):135‐143.

Mwendera,E.J.,1994.EffectonthewateryieldoftheLuchelemucatchmentinMalawiofreplacingindigenousgrasseswithtimberplantations.ForestEcologyandManagement,65(2‐3):75‐80.

Oki,T.,Kanae,S.,2006.Globalhydrologicalcyclesandworldwaterresources.Science,313:1068‐1072.Pagiola,S.,2008.PaymentsforenvironmentalservicesinCostaRica.EcologicalEconomics65,713–725.Pattanayak, SK., 2004. Valuing watershed services: concepts and empirics from Southeast Asia. Agriculture,

EcosystemsandEnvironment,104:171‐184.

46

Raghunath, B., Das, D. C., Thomas., P. K., 1970. Some results of investigations on hydrology of the sub‐watersheds in the Nilgiris (India). In: Proceedings of Symposium on the Results of Research onRepresentativeandExperimentalBasins,Wellington,NZ.,IAHSPublicationnº96,pp.416‐440.

Sahin,V.,Hall,M.J.,1996.Theeffectsofafforestationanddeforestationonwateryields.Journalofhydrology,178:293‐309.

Samraj, P., Sharda, V.N., Chinnamani, S., Lakshmanan, V., Haldorai, B., 1988. Hydrological behaviour of theNilgirisub‐watershedsasaffectedbybluegumplantations,PartI.Theannualwaterbalance.JournalofHydrology,103(3‐4):335‐345.

Scott,D.F.,Bruijnzeel,L.A.,Mackensen,J.,2005.Thehydrologicalandsoilimpactsofforestationinthetropics.In:M.BonellandL.A.Bruijnzeel(eds.),Forests,WaterandPeopleintheHumidTropics:Past,presentand future hydrological research for integrated land and water management. Cambridge UniversityPress,pp.622‐651.

Scott,D.F.;Bruijnzeel,L.A.;Vertessy,R.A.;Calder,I.R.2004.Foresthydrology:impactsofforestplantationsonstreamflow. In: J. Burley, J. Evans and J.A. Youngquist (eds.), The Encyclopedia of Forest Sciences.Oxford,Elsevier,pp.367‐377.

Shadish,W.R.,Haddock,C.K.,1994.Combiningestimatesofeffectsize. In:H.Cooper,L.V.Hedges(eds.).TheHandbookofResearchSynthesis.RussellSageFoundation,NewYork,pp.261‐281.

Sharda, V. N., Samraj, P., Samra, J.S., Lakshmanan, V., 1988. Hydrological behaviour of the Nilgiri sub‐watersheds as affectedby blue gumplantations, Part II.Monthlywater balances at different rainfallandrunoffprobabilities.JournalofHydrology,103(3‐4):347‐355.

Sikka,A.K.,Samra,J.S.,2003.Lowflowandhighflowresponsestoconvertingnaturalgrasslandintobluegum(Eucalyptusglobulus)inNilgiriswatershedsofSouthIndia.JournalofHydrology,270(1‐2):12‐26.

SmakhtinV.U.,2001.Lowflowhydrology:areview.JournalofHydrology240:147‐186.Susswein,P.M.,vanNoordwijk,M.,Verbist,B..,2001.Forestwatershedfunctionsandtropicallandusechange.

Towardsintegratednaturalresourcemanagementinforestmarginsofthehumidtropics: localactionand global concerns. ASB Lecture Note 7. International Centre for Research forAgroforestry, Bogor,Indonesia,38pp.

Tognetti,S.S.,Mendoza,G,Aylward,B.,Southgate,D.,Garcia,L.,2004.Aknowledgeandassessmentguidetosupportthedevelopmentofpaymentarrangementsforwatershedecosystemservices.WorkingPaper.EnvironmentDepartment,WorldBank,Washington,DC.

UNESCO, WMO, 1992. International glossary of hydrology. Paris, United Nations Educational Scientific andCulturalOrganization;Geneva,WorldMeteorologicalOrganization,1992.

vanNoordwijk,M.,Poulsen, J.G.,Ericksen,P.J..,2004.Quantifyingoff‐site effectsof land‐use change: filters,flowsandfallacies.Agriculture,EcosystemsandEnvironment,104:19‐34.

Vanclay, J.K.,2009.Managingwateruse fromforest plantations.ForestEcologyandManagement257:385–389.

Vertessy, R.A., Watson, F.G.R., O'Sullivan, S.K., 2001. Factors determining relations between stand age andcatchmentwaterbalanceinmountainashforests.ForestEcologyandManagement,143:13‐26.

Waage, S., Bracer, C., Inbar,M., 2008. Payments for EcosystemServices:Getting started. Forest Trends, theKatoombaGroup,UNEP,WashingtonDC.

Wilk,J.,Andersson,L.,Plermkamon,V.,2001.HydrologicalimpactsofforestconversiontoagricultureinalargeriverbasininnortheastThailand.HydrologicalProcesses,15(14):2729‐2748.

Wunder, S., 2005. Payments for environmental services: some nuts and bolts. Occasional Paper 42, CIFOR,Bogor,Indonesia,32pp.

Wunder,S.,Albán,M.,2008.Decentralizedpaymentsforenvironmentalservices:ThecasesofPimampiroandPROFAFORinEcuador.EcologicalEconomics,65:685‐698.

Wunder, S., Engel S., Pagiola S., 2008. Taking stock: A comparative analysis of payments for environmentalservicesprogramsindevelopedanddevelopingcountries.EcologicalEconomics65:834‐852.

47

Zhou, G.Y., Morris, J.D., Yan, J.H., Yu, Z.Y., Peng, S.L., 2002. Hydrological impacts of reafforestation witheucalypts and indigenous species: a case study in southernChina. Forest Ecology andManagement,167(1):209‐222.

49

3. Decision making by farmers regarding ecosystem services: factors affecting soil conservation efforts in Costa Rica

Paperpublished:VignolaR.,Koellner,T.,ScholzR.W.,McDanielsT.L.,2009.Decision‐makingof

upstream farmers for targeting policies on ecosystem services: cognitive and socio‐economic

variablesforsoilregulationservicesinCostaRica.LandUsePolicy27,1132‐1142

Abstract

Theimpactofclimatechangeonfarmsoilsinthetropicsisthecombinedresultofshort‐termsoil

management decisions and expanding precipitation extremes. This is particularly true for

cultivated lands located in steeply slopingareaswherebare soil isexposed toextreme rainfall

suchastheBirriswatershedinCostaRica.Farmersinthiswatershedareaffectedby increasing

degradation of soil regulation services and respondwith different level of efforts to conserve

their soils. This paper examines influences on farmers’ decisions through a survey involving

interviews with a sample of farmers (n = 56) to test hypotheses on how a combination of

cognitivevariables(beliefs,riskperception,values)andsocioeconomicvariablesshapedecisions

onsoilconservation.Resultsshowthatfarmers’awarenessoftheirexposureleveltosoilerosion

combineswithothervariables todetermine their level of soil conservation.Usingdiscriminant

analysis,threegroupsoffarmerswereidentifiedbasedontheirsoilconservationefforts.ANOVA

pairwise‐comparisonamong thesegroups showedsignificantdifferences in respect to levelsof

awareness, perception of risk, and personal beliefs along with territorial exposure and

participation in soil conservation programs. Our results help to understand farmers’ complex

decision‐makingonsoilconservationandhelpdesigningpoliciestosupporttheprovisionofsoil

regulation services especially in areas highly exposed to increasing frequency of extreme

precipitationeventssuchasCentralAmerica.

Keywords:Soilregulationservices,cognitivevariables,Farmersdecision‐making,CostaRica

3.1Introduction

Soil erosion, land use and water management are highly interlinked. In Central America,

unsustainableuseofmarginal lands iswidespreadandhascausedseveresoilerosion,which in

turn reduces productivity of soils in upstream areas (Pimentel et al., 1995) and, downstream,

50

water quality regulation services (Southgate andMacke, 1989; Lutz et al., 1994a; Guo et al.,

2000). Inaddition to theseanthropogenic factors,precipitationdistributionplaysan important

roleinerosion.AccordingtothemodelofWischmeierandSmith(1978),extremerainfallevents

interact with high sloping and unprotected soils, increasing kinetic potential of raindrops to

remove soilparticles increasingerosionand sedimentationdownstream.ForCentralAmericaa

significantincreaseinintensityandfrequencyofextremeprecipitationeventshasbeenobserved

(Aguilaretal.,2005)andisprojectedtoincreasefurther(Magrinetal.,2007).

Exposure to precipitation extremes and the high susceptibility of soils to erosion due to

questionablemanagementmakethisregionhighlyvulnerabletoclimatechange.TheCostaRican

MinistryofEnvironmentestimatedthatreducedsoil fertilityandsoilerosion,partlyduetothe

human and natural factors described above, caused a 7.7% reduction in agricultural Gross

Domestic Product from 1970 to 1989 (MINAE, 2002). Moreover, the National Commission on

LandDegradation(CADETI,2002)estimatesthat20%ofagriculturallandisseverelyover‐utilized

due topoormanagementpracticesasa resultof short‐term responses toproductiveneedsof

farmers.

Policies and development programs to foster soil conservation in vulnerable watersheds of

tropicalcountries likeCostaRicausetools likedirecteconomic incentivessuchasthePayment

for Ecosystem Services (PES) scheme (Pagiola, 2008;Wunder et al., 2008). However, adoption

rates of conservation practices in Costaricanwatershedswhere intensive andmarket‐oriented

agriculturalactivitieshavecausedlargedeforestationandsoilsarepronetointenseerosionare

low (Abreu, 1994). This is the case of the Birris watershed where other factors than direct

payments are influencing farmers proneness to adopt soil conservation practices. Here, PES

schemesneedtocomplementexistingdirectincentiveswithotherstrategiestoachieveintended

additionality.Ananalysisofinstitutionalandcognitiveaspectsassociatedwithsoilconservation

cangiveinsightstothedesignofsuccessfulprograms(SWCS,2003).

Thereasonforthatisthatoftenthecost‐effectivenessofsoilconservationprogramsdependson

whether erosion is considered only for its off‐site effects or also for on‐site effects (i.e. in the

lattercasefarmersmighthaveadirectpersonalincentivetoimplementvoluntaryconservation)

(WossinkandSwinton,2007;DaleandPolanski,2007).Soilconservationprogramsshould,then,

use complementary tools to foster adoption of adequate practices among farmers. This paper

51

intendtoprovideinputstothedesignofappropriatesoilconservationprogramwhileanalyzing

thecomplexityofsocioeconomicandcognitivevariablesthatinfluencefarmers’decision‐making

forsoilconservation.

The objective of this paper is to analyze how cognitive and socioeconomic variables influence

farmers’decisionsregardingtheirmanagementofsoils.Morespecifically,wetesthypotheses,in

ourcasestudy,concerningtheinfluenceofcognitive,exposure,andsocioeconomicvariableson

farmers’ adoption of soil conservation practices.We test the following null hypotheses: i) risk

perceptionofnaturalandanthropogenicactivities,andknowledgeofthefactorscausingerosion;

ii) traditional socioeconomic variables (farm size, tenure, education, economic status); and iii)

landcharacteristics,suchaslocationinhighriskareas,consideredindividuallyandtogether,have

noeffectonsoilconservationdecisions.

This paper is organized in the following manner. Section 2 discusses the decision‐making

processesoffarmersrelatedtosoilconservationreferringtopreviousliteraturefindings.Section

3 presents the case study, and describes themodel and themethods. Section 4 presents the

results in terms of a series of analytical steps and the construction of cognitive variableswith

factoranalysis.Section5providesdiscussion,followedbyconclusionsinSection6.

3.2Farmers’Decision‐MakingforSoilConservation

A large body of literature has analyzed farmers’ decision‐making to inform the design of soil

conservationprograms.Econometric studiesbasedon farmers’monetaryutility functionshave

beenwidelyused to studyadoptionof soil conservationpractices (FeatherstoneandGoodwin,

1993;InnesandArdila,1994;Souleetal.,2000;IllukpitiyaandGopalakrishnan,2004;Leyvaetal.,

2007).Neo‐classicaldecisionmodelsassumingmaximizationhavebeencritizedbysomewriters

as inappropriate for modeling behaviour in real world decisions since strong assumptions on

cognitiveaspectsarerequired(e.g.accesstoperfectinformation,influenceofsocialcontext,and

individuals’andsocialpsychologicalcharacteristics) (VanderBerghetal.,2000).Theseauthors

alo suggest that especially in developing countries, a changed behavioural economics that

includesecologicalandpsycho‐socialdimensionsofagricultureisemerging.

Societal responsetoreduce increasingvulnerabilitytoerosion requiresphysicalassessmentsof

erosionprocessesbutalsoananalysisof institutionalandcognitiveaspectsassociatedwiththe

52

problemanditssolution(SWCS,2003;GrothmannandPatt,2005;Parryetal.,2007).Thereason

for that is thatoften thecost‐effectivenessof soil conservationprogramsdependsonwhether

erosion isperceivedonly for itsoff‐siteeffectsoralso foron‐siteeffects (i.e. in the latter case

farmers might have a direct incentive to implement voluntary conservation) (Wossink and

Swinton,2007).Indeed,actionsdirectedtoimprovesoilregulationhavethepotentialtobenefit

other societal sectors. For example, soilmanagement practices such as using low soil‐removal

ploughscanreducesoilerosionbutalsoincreasewaterinfiltration(Swintonetal.,2007;Daleand

Polanski,2007).

Whenstudyingfarmers’conservationbehaviour, theperceptionofuncertaincostsandbenefits

ofconservationpracticesisalsorelevant.Asprospecttheory(TverskyandKahneman,1974)and

other theories indicate, decision‐making under uncertainty is influenced by cognitive aspects

such aswhat is considered a loss or gain, experience, individual values and a series of beliefs

aboutthefunctioningofthesysteminwhichdecisionsaremade.

Tostudyfarmers’voluntaryconservationefforts,Gouldandcolleagues(1989)usedathreephase

decision model: identifying the problem, deciding on adoption and determining the level of

effort.Theseauthorsfoundthat,forsocioeconomicvariables,theeffectoffarmsizeonadoption

of soil conservation practices was negative while that of income was positive. In the case of

cognitivevariables, theperceptionof severityof impactsof soil erosionwaspositive. Similarly,

Lynne et al. (1988) combined psychological variables measuring attitudes, with economic

variables, such as income and tenure. In theirmodel, attitudeswere used as an expression of

expectedvalueattachedtoagivenconservationactivity.Theseauthorsfoundthatcognitiveand

psychologicalvariablessuchasbeliefsonconservationofand responsibilityonmanagementof

soilhadapositiveeffectonadoptionofsoilconservationpractices.

Along the same line, theon‐siteperceivedeffectsof conservationbehaviourwereanalyzedby

Traoreetal.(1998),whoconfirmedthatperceptionofenvironmentalproblemsbyfarmerswas

correspondedbyahigheradoptiondegreeofenvironmental‐friendlyagriculturalpractices,while

socioeconomicvariablesliketenureandfarmsizewerenotsignificant.

The role of risk perception, beliefs, values and socioeconomic variables in farmers’ decision‐

making on soil conservation has been a concern of environmental policies to address climate

changethreats.Riskperception,environmentalvalues(cost‐benefitexpectancy)andbeliefs(i.e.

53

knowledge base) have different explanatory power for decisions that address environmental

change(O’Connoretal.,1999;SiegristandGutscher,2006).Forexample,LeeandZhang(2005)

used psychometric measurements to analyze farmers’attitudes and risk perceptions of land

degradationalongwithsocioeconomicvariables.Theseauthorsmeasuredtheriskperceptionsof

farmersfortheeffectsofgrazing,openingupfarmland,cuttingtreesforfuel‐woodanddigging

upalocaltraditionalplantonlanddegradation.Typeofoccupationandagehaveapositiveeffect

on awareness of land degradation and on the perception of its impacts. Farmers perceive the

existence of different ecological risks associated with land degradation activities although

significant misconceptions on land degradation causes and impacts exist among farmers’ and

experts’knowledge.

Another approach, utilized by Bayard and Jolly (2007), adapted the Health Belief Model as a

modifiedversionofAjzen’stheoryofPlannedBehaviour(1991).Intheirapproach,psycho‐social,

physical and economic constructs were used to measure persons’ beliefs about costs and

benefitsofagivenbehaviourwhichdetermine intentionandsoilconservationbehaviour.More

specifically, four components associated with the risk of soil degradation were included:

perceived susceptibility (e.g. exposure); perceived severity (e.g. size of impacts); perceived

benefits(privatevspublic),andbarriers(e.g.perceivedbehaviouralcontrol).Thefindingsconfirm

that awareness of the problem has a positive effect on soil conservation behaviour, while

attitudes had no significant effect. On the other side, awareness is positively influenced by

farmers’higherperceptionofseverityandsusceptibilityofsoildegradationrisk.

Explicitly accounting for climate change risk, Grothmann and Patt (2005) used the Private

Proactive Adaptation to Climate Change decision model. In this model, farmers’ access to

resources,institutionalbarriers,experience,theperceptionofprobabilityandseverityofclimate‐

related events and, finally, perceived capacity to respond were determinants in farmers’

adoptionofresponsivebehaviour.Theseauthorsappliedqualitativelytheirmodeltothecaseof

farmers’ adoption of climate forecast information in their farming decisions. They could show

that there is inconsistency among farmers’ and experts’ assessment of risk and that low

perceived self‐adaptive capacity hinder adoption of risk‐preventive behaviour. In Central

America, farmers’ soil conservation decisions have been studied mainly from a traditional

54

utilitarianperspective,withsomeexceptionsfoundintherecompilationofCentralAmericacase

studiesoneconomicandinstitutionalanalysisofsoilconservationprojects(Lutzetal.,1994).

Noneoftheseregionalstudiesanalyzedtheinfluenceofriskperceptionsandbeliefsconcerning

anthropogenic activities and natural events as determinants of preventive behaviour of soil

erosioncontrol.Considering theconcernsof the international communityandnationalpolicies

onreducingthepotential impactsofclimatechange, it is relevanttoexplorehowandwhether

riskandcausesofsoilerosionareperceivedandwhethertheseinfluencefarmers’conservation

effortsinahighlyvulnerableregion.

3.3Methods

3.3.1Descriptionofthearea

TheBirris (coordinates 9.9o N, 83,8oW) is a sub‐watershed of the ReventazonRiver in central

CostaRicaflowingintotheCaribbean,andis4800hectaresinsize.Thewatershedisinfluenced

by the Caribbean climate, with 2325 mm average annual rainfall of which about 80% is

concentratedintheperiodfromMaytoDecember.Thetopographyischaracterizedbyslopesof

up to 70 %, especially in the upper part of the watershed. The population density is 161

inhabitantspersquarekilometer(i.e.abovethenationalaverage;INEC,2002);mostinhabitants

arelocallybornandthemajority(61%)areinvolvedinagriculture(ICE,2000).

Large‐scalelandconversionofthecloudforestintheBirriswatershedstartedinthe1960swhen

horticulture and dairy‐cattle pastureswere established. Intensive soil use for horticulture (217

producersofpotato,cabbageandcarrotswith2.5haaveragefarmsize)anddairy‐cattlepastures

(122producerswith5to7haaveragefarmsize), is foundmainly inriparianareas.Agriculture

andinfrastructureshavereducedthecoveroftropicalcloudforestto28%ofthetotallandarea

in the watershed. This intense process of forest fragmentation and intensive agricultural

production makes this area one of the largest sediment‐producers in the country (Sanchez‐

Azofeifaetal.,2002).Averageerosionratesgrewfrom12ton/ha/yrpriorto1978,whenonly15%

ofthewatershedwasunderhorticulture, to42ton/ha/yr in1992(Abreu,1994).However, the

effect of these high levels of erosion is only visible in some areas3, since deep andosols are

3Thatis,probablyonlysomefarmersarestartingtoperceiveitsdirectimpactsoferosion.

55

common inthearea(Lutzetal., 1994b;MarchamaloandRomero,2007).TodatetheNational

ElectricityInstitute(ICE)yearlyspendsmorethanfourmillionUS$tocleanitsdamsandaround

threehundredthousandUS$topromotethesoilwatershedmanagementplan.Ontheotherside

theministryofagricultureprovideshumanand logistic resources topromote soil conservation

practicesinupstreamareas.Here,arecentestimationoftherepositioncostofsoilnutrientslost

witherosioncorrespondstoninety‐fivethousandsUS$ayear(Vignolaetal.,2010).

3.3.2SampleandSurveyInstrumentDesign

To develop the survey instrument we conducted two focus groups and interviews with key

informants(farmersandagriculturalextensionofficers)toidentifyaspectstobeincludedinthe

analysisoffarmers’adoptionofsoilconservationpracticesintheBirriswatershed.Additionally,

we also reviewed instruments applied in previous farmers’ adoption studies that analyzed

attitudesandvaluesrelatedtosoil conservation (seesection3.3).Wereviewedquestionswith

agriculturalextensionofficerstoensurethattheywouldbeeasilyunderstoodbyfarmers.Apilot

testwasconductedwithfivefarmerstofurtherclarifyquestions.

Oursample(n=56;N=300)iscoveringaround18%ofallfarmersinthewatershedandissimilar

to that of previous studies similar studies (n = 64; Bewket, 2007). Three lists of producers

availablefromthelocalagricultureextensionofficewereusedforastrata‐proportionalselection

offarmerscorrespondingtothreecategoriesofproducerspresentinthewatershed:horticulture

anddairy‐cattle ( )10n = ,horticultureonly ( )35n = anddairy‐cattleonly( )11n = .Producerswere

visited in the field or at home andwere administered by pre‐trained interviewers a half‐hour

structured interview consisting of the following sections: i) soil conservation practices

implemented; ii)perceptionof farmsoilquality; iii) risk factors contributing to soil erosion; iv)

values that influence soil conservation decisions; and, finally v) questions regarding socio‐

economicdata.

3.3.3TheDecisionModel

Building on Bayard and Jolly (2007), we propose the followingmodel of farmers’ decisions to

adoptsoilconservation:

( ), , ,CD f B RP V SE= ,

56

wheresoilconservationdecisions ( )CD dependonthreetypesofindependentvariablessuchas

theirbelief/knowledge( )B , theirriskperceptions( )RP ,values ( )V andasetofsocio‐economic

characteristics ( )SE .Thismodelservedasthebasisoftheresearchdesigndiscussedbelow.

To measure farmers’ soil conservation behaviour we used secondary literature and key

informants (e.g.agriculturalextensionofficersandmembersof farmersassociations)tobuilda

list of conservation practices that are promoted by existing extension programs or by farmers

themselves in thewatershed (ICE,2000;Marchamalo, 2004).The listof conservationpractices

includes planting of trees, cultivation along contour lines, use of water and soil conservation

channels,naturalregenerationofvegetation incriticalareas,amongothers. Weconstructeda

binaryvectorforeachpracticebyaskingrespondentstomarkthosepracticesonthelistthatthey

wereimplementingwhentheinterviewswereconducted.

We then constructed for each farmer i the dependent categorical variable soil conservation

effortCE(i) = 1 forloweffort,CE(i) = 2 formediumeffort,andCE(i) = 3 forhigheffort.Todo

this,wefirsthavetoconsiderthattherearedifferentnumbersofpossibleprotectionpractices

Pj(i) foreachproductivecategory j ( j = 1 forcattlebreeding; j = 2 forhorticulture, j = 3 for

bothpractices).Forafarmer i belongingtoproductioncategory j ,wecalculatetheconservation

score:

YConsj(i) =

P(i)

Tj

,

where P(i) is the total number of practices implemented by the farmer i and Tj is the total

numberofpossiblepracticesforproductivecategoryj.Theconservationscorevariablewasused

to define CE(i) by: (1)CE if less than half of the protection practices are applied (i.e. if

YConsj(i) < 0.5 ); (2)CE ifmore thanhalf andup to75%of theproductionpracticesareapplied

(i.e. if 0.5 ! YConsj(i) ! 0.75 );and (3)CE ifmorethanthree‐quartersofallpossiblepracticesare

applied(i.e.if 0.75 < YConsj(i) ).

Cognitivevariables

57

Wedrawonsimilarstudies(Bewket,2007)thatusedexplanatoryvariablesoffarmers’decisions

toimplementsoilconservation,suchasawarenessandperceptionoferosionhazard,itscauses

andcontrollability.Inthissection,weexplorekeyconceptsunderlyingthesecognitivevariables.

Farmers’beliefs ( )B guidetheirunderstandingandevaluationofcausesandsolutionsassociated

witherosioncontrol(D’emdenetal.,2008).WefollowedtheapproachofWagner(2007),who

usedmentalmodelquestionstoelicitlay‐people’sbeliefsoncausesofflashfloods,andBewket

(2007) who asked farmers about their beliefs on causes of soil erosion. Rankings on the

importance of five degradation activities reported by recent studies (ICE, 2000; Marchamalo,

2004; Ramirez, 2008) were used, namely: deforestation, agricultural practices, extreme

precipitation,urbanization,anddairycattlepastures.Wealsoaskedquestionsonwhetherthey

perceive that the “general health”4 of their soil has changed from the past andwhether they

think itwill change in the futureasaway tomeasure their judgmenton thecontinuityof the

erosion problem over time. Additionally, perceptions on the positive role of trees in erosion

controland riparianprotection,as recognized inthe literature(Tejwani1993;Bruijnzeel,2004)

andpromotedbysoilconservationprograms,weremeasured.

Risk perception ( )RP was measured following the psychometric paradigm which allows

identification of differences in perceptions among groups of individuals (Slovic, 1987). In this

paper, risk perceptionmerely denotes the cognitive representation and evaluation of negative

impacts resulting from a given activity that changes the environment. Risk of erosion can be

definedasconsequence‐relatedriskthat,usingtheclassificationofSlovic(1987),canbejudged

observable, is relatively well‐known to scientists, has immediate effect and, considering the

highlyexposedcontextofour study, is a relativelywell‐knownandoldproblem in thearea. In

thisstudy,theidentificationofriskconstructswasadaptedfrompreviousstudiesthatanalyzed

the relationship among environmental risk sources and behavioural intentions and responses

(McDaniels et al., 1996; McDaniels et al., 1997; O’Connor et al., 1999; Lee and Zhang, 2005;

BayardandJolly,2007).

4 This concept was used by key informants when discussing on soil erosion. Soil erosion is associated by farmers to theconcept of decreasing soil health which they perceive associated to several consequences such as decreasing productivecapacityandincreaseinamountoffertilizersneeded.

58

Farmers’ decisions regarding soil conservation are largely driven by short‐term perceived risks

associated with markets and weather, rather than long‐term climate change‐related hazards

(GrothmannandPatt,2005).Tomodelthiscomponent,andfollowingBayardandJolly(2007),we

builtconstructstomeasureriskperceptionusingLikertscalesfromonetoseventomeasurethe

individuals’perceptionoftheimmediacyandsaliencyofimpactsfromdriversofsoilerosionsuch

as agriculture, cattle, rainfall, and logging and urbanization. This immediacywasmeasured by

four constructs; namely, severity, uncontrollability, observability and rapidity. Additional

questionswere intendedtomeasuretheextenttowhich individualsperceivethemselvestobe

livinginareasgenerallyexposedtosoilerosionrisk.

Values ( )V characterize those standards that individuals judge appropriate and desirable in a

givensituation,andthusreflecttheirorderingofpreferencesforend‐statesoftheenvironment

(Scholz,n.d.).Webuild,forthevalueitems,onLeeandZhang(2005)andO’Connoretal.(1999)

to formulate questions on how much farmers and public administration should be held

responsible for avoiding damage to soil. Finally, all items used to measure all our cognitive

variablesarepresentedintable1.

59

Table 1: Items and their codes to measure cognitive variables. Component Abbreviations Items

BProdInRipAgua Riparianareasareoptimalformyproductiveactivitiesgivenproximitytowater.BProdRipGood Slopingareasaregoodformyproductiveactivity.BSoilConsCostInco Soilconservationmeasuresdon'tbringmuchbenefit,itismoretheinvestmentcostthen

thereturnsformyfarm.BSoilNeverDegr Soilisaninexhaustibleresource,whichiswhyitcanbeexploitedcontinuously.BScienceTech Scientificandtechnologicalprogressinsoilmanagementallowsovercomingsoil

degradationproblemsBSoilHealth* Howhealthydoyouthinkyoursoilis?BSHchange** Doyouconsiderthatyoursoilhealthhaschangedcomparedtothepast?BSHChFut Doyouthinkyoursoilhealthmightchangeinthefuture?BContrPreci Howmuchdoesrainfallcontributetoerosion?BContrAgr Howmuchdoagriculturalpracticescontributetosoilerosion?BImpoRain Amongdeforestation,agriculture,rainfall,roadsandcattlebreeding,howwouldyourank

precipitationimpactsonsoilerosion?BImpoAgr Amongdeforestation,agriculture,rainfall,roadsandcattlebreeding,howwouldyourank

theimpactsofagriculturalactivitiesonsoilerosion?BTreesDecrProd Treesonfarmlandsdecreaseproduction.BTreeConsSoil Treesbenefitsoilconservation.

Belief

BSlopeFor Inhighslopingareasthereshouldbenoproductiveactivity,onlyforest.RPCattleObs Howobservableistheimpactofcattleonsoilerosion?RPCattleRap Howfastistheimpactofcattleonsoilerosion?RPAgrRap Howfastistheimpactofagricultureonsoilerosion?RPCattleSev Howlargeistheimpactofcattleonsoilerosion?RPAgrObs Howobservableistheimpactofagricultureonsoilerosion?RPAgrSev Howlargeistheimpactofagriculturalactivitiesonsoilerosion?RPEroPast Howmuchhaserosionriskincreasedinrespecttothepast?RPEroFut Howmuchwillsoilerosionriskincreaseinthefuture?RPCrSev Howlargearesoil‐erosion‐relatedlossesinCostaRica?RPRainObs Howobservableistheimpactofprecipitationonsoilerosion?RPRainSev Howlargeistheimpactofprecipitationonsoilerosion?RPRainRap Howfastistheimpactofprecipitationonsoilerosion?RPCrContr HowmuchcanpeopleinCostaRicacontroltheriskofsoilerosion?RPRainContr Howcontrollableistheimpactofprecipitationonsoilerosion?RPCattleContr Howcontrollableistheimpactofcattleonsoilerosion?

Risk

RPAgrContr Howcontrollableistheimpactofagricultureonsoilerosion?VSoilProdNoDestr SoilproductivityshouldneverbedestroyedVwstrUseCare Downstreamusersdon'tcareabouthowweupstreamproducersmanageourfarmsoilsVIncomeNow Ensuringmaximumreturnpossibleforthisyearismuchmoreimportantthanensuringsoil

productivityforfuturegenerationsVProdPay Producerswhoareresponsibleforsedimentationinriversshouldpayforremovingsuch

sedimentsfromriversanddamsVProdOwnDec EachproducerownshisfarmsoilsothathissoilmanagementdecisionsarestrictlypersonalVRespFutGenr Forourchildren,weproducershavetheresponsibilitytoreduceerosion

Values

VPublicAdmPay Publicadministrationresponsibleforroadsshouldpayforremovingsedimentsproducedbythisinfrastructurefromriversanddams

AllvariablesaremeasuredonLikert‐scales.Beliefsscalesare1(donotagree)to7(verymuchagree);except*(veryunhealthy=1toveryhealthy=4)and**(Hasworsened=1,thesameasnow=2,hasimproved=3);Riskperceptionsarefrom1(verylittle)to7(verylarge);Valuesscalesare1(donotagree)to7(verymuchagree).

Socioeconomicvariables

As reported in other studies (Grasmuck and Scholz, 2005; Siegrist and Gutscher, 2006) risk

perception and its associated responsive behaviour is influenced by individuals’ exposure.We

60

createdabinaryvariabletodiscriminateproducersbyriskareasthatwereidentifiedbypolygons

(see Figure 1) reflecting the boundaries of communities subject to the highest rates of soil

erosion(Marchamalo,2004).

Figure 1: Location, current land uses and the delimitation of areas with high soil erosion of the Birris watershed, (A: Costa Rica; B: Reventazon watershed; C: Birris watershed).

Existingprogramspromotingsoilconservation intheBirriswatershedcovera limitedextentof

the area. To account for their eventual influence on individuals’ judgment and conservation

behaviour,participation inanexistingprogramwasmeasuredwithabinaryvariable.Economic

welfarelevelwasincluded,similartootherstudies(Lynneetal.,1988;O’Connoretal.,1999;Lee

and Zhang, 2005; Bayard and Jolly, 2007), with a ranking variable of monthly household

consumption with seven levels. As seen in section 2, land tenure has widely been used in

studyingfarmers’adoptionofsoilconservationpractices.So,basedonourrecollecteddatawe

constructedanownershipvariable(“owned”)whichmeasuresthepercentageoftotalfarmland

61

thatisownedbytheindividualproducer.Finally,educationlevel(measuredby5levels)andage

of individualswere included for their important role outlined inprevious research on farmers’

conservationbehaviour(Traoreetal.,1998;IllukpitiyaandGopalakrishnan,2004;LeeandZhang,

2005).

3.3.4Statisticalanalysis

3.3.4.1Constructionofcognitivevariables

In what followswe describe howwe used factor analysis in SPSS (2001) on B , RP , and V to

create three indexes thatwereultimatelyused, togetherwith socioeconomicvariables, to test

our hypotheses. The three sets of indexes ( ), ,B RP V were obtained by conducting the factor

analysis with the Kaiser normalization method, varimax rotation, and Principal Component

Analysisastheextractingmethodwiththevariablesundereachheadingintable1.Onlyfactors

withaneigenvalueof1orabovewereretained(Kaiser‐Gutmannretentioncriterion;Kinneletal.,

2002).Fitness for factorialanalysiswastestedwithBartlets’KMOand p value! .Rotatedfactor

loadings were used to identify relevant variables to be used in the construction of indexes.

Variableswereretainediftheirrotatedloadingswereabovethethresholdvalueof0.5.

In the exploratory analysis, we excluded from further analysis the variable that measured

farmers’ judgmenton“own responsibilitynot todestroy soil productivity” ( )SoilProdNoDestrV for its

low variance (ratings between 5 and 7), possibly due to strategic answering. Then, a separate

factor analysis5 (i.e. one for each specific cognitive aspect of the decisionmodel) reduced the

numberofvariablesfromtheoriginal38to9cognitiveindexes.Ineachfactoranalysisweutilized

standard procedures such as KMO values and commonalities to improve the separation of

factors.Thefinalresultsareshownintable2.

Table 2: Socio-economic variables of the sample (n = 56). Variable Average S.D. Min Max

Age(years) 48.98 14.55 18 80

Farmsize 18.30 11.11 1 38

Educationlevel 2.29 0.87 1 5

5Inanattempttoreducethenumberofvariables,weranexploratoryfactoranalysesusingallvariablesmeasuringcognitive

aspects.Thelowfitness‐testvalue ( )0.352; 0.0001KMO p= < ,however,didnotseparateconsistentfactorsoutofthe

14loaded,thussuggestingtheneedforaminor‐aggregationanalysis.

62

Monthlyconsumptionlevel 4.45 1.37 2 8

Beliefs. The factor analysis on beliefs variables identified 3 underlying

factors( )0.671; 0.01KMO p= < . With factor 1 loadings, we constructed an index1B called

“Positive Thinking” ( )1 ProdInRipAgua SoilNeverDegr ScienceTechB B B B= + + composed of items capturing i)

beliefsthatsoilsareresilienttodegradation, ii) thattechnologicalprogress insoilmanagement

practicesissufficienttocountersoildegradationand,finally,iii)thatproducinginriparianareas

(i.e. sloping areas) is good because water is closer. This index thus indicates the triumph of

positive thinkingaboutproductionand technologyoverworries on soildegradationprocesses.

Items from factor 2 loadings built index2B named “soil critical conditions”

( )2 SoilHealth BSHchangeB B= + .Theseitemsmeasurebeliefsastohowhealthyfarmersthinktheirfarm

soil isandwhether ithaschanged.Thescalesofthetwocontributoryitemswerestructuredto

measure how strongly they believe their farm soil to be in critical condition (i.e. higher values

correspond to higher awareness of critical soil conditions). Factor 3 loadings identified beliefs

associatedwiththeimportanceofrainfallinerosionprocessesingeneralandintheirfarmsoilin

particular. Index3B , termed “Climate extremes” ( )3

ContrPreci ImpoRainB B B= + , captures beliefs

associatedwiththiscomponentoftheerosionproblem.

Risk perception. Factor analysis for risk perception ( )0.71; 0.0001KMO p= < identified four

factors. Factor 1 loadings identified an underlying index1

RP called “Agricultural

risk”( )1 AgrContr AgrObs AgrSev AgrRapRP RP RP RP RP= + + + , which captures risk perception associated

with agricultural activities. More specifically, it measures the immediacy of their impacts on

erosion and how controllable they are and tells us about the relevance of these activities for

erosion.Basedontheloadingsofthesecondfactor,weconstructedindex2

RP ,termed“Pasture

risk”, ( )2 CattleObs CattleSev CattleRapRP RP RP RP= + + , representing the perception associatedwith the

immediacy and severity of pasture activities on erosion. The index 3

RP

( )3 EroPast RainObs RainRapRP RP RP RP= + + isnamed“ExtremePrecipitation”andclearlycomplements

the first two risk perception indexes, which are directed at anthropogenic activities, by

63

addressingjudgmentonnaturalphenomena.Thisindexincludesperceptionoftheimmediacyof

rainfallimpactswhich,interestingly,iscorrelatedinthesamefactorwiththeperceptionofhow

strongly erosion has increased in respect to the past. Finally, in the factor analysis of risk

perception,thefourthfactorloadeditemsmeasuringperceptionofthecontrollabilityoferosion

effectstogetherwiththeextenttowhichindividualsperceivethatpeopleinthecountryareable

to control this environmental problem. We name index4

RP “control

perception”( )4

CrContr RainContrRP RP RP= + since it measures perceptions of how controllable the

erosion effects of extreme precipitation are, alongwith the capacity of Costa Rican people to

copewiththem.

Values. Factor analysis on values items identified two underlying

factors( )0.559; 0.05KMO p= < . The first factor loaded items measuring values on private

aspects of the erosion problem. Thus, this index called “private

benefits”( )1 IncomeNow ProdOwnDec ProdNoPayV V V V= + + measures preference for short‐term benefits,

right to decision on their own farm soil (i.e. erosion is not of public concern) and, finally, the

opinionthatproducersshouldnotbepayingforremovingsedimentsfromtheriver.Thesecond

factor, in contrast, identified an index ( )2 VRespFutGenr PublicAdminPayV V= + that captures the values

related more to the public than to the private benefits. This index is built with variables

measuringconcernsonwhetherpublicadministration(i.e.andnotindividualfarmers)shouldbe

held responsible for removing sediment from rivers and whether soil conservation should be

ensuredforthebenefitsoffuturegenerations(i.e.expressingconcernsforlonger‐termbenefits

oferosioncontrol).

3.3.4.2Differentiationoffarmers’types

Inasecondstep, followingCarlssonetal. (1977),weuseddiscriminantanalysis,using ( )CE i as

thedependentvariables,toprojectthesocioeconomicandcognitiveindexesinabi‐dimensional

space allowing for visual representation of the farmers’ conservation‐efforts groups and the

variables characterizing them. Finally, we tested our hypotheses in section 1with a pair=wise

ANOVAtocomparegroups’meansscoresoncognitiveandsocioeconomicvariables.

64

3.4Results

3.4.1Sampledescription

Most farmers in our sample are full‐time agricultural producers (63%) and have relatively low

educationallevels(i.e.primaryschool).Theaveragelevelofhouseholdmonthlyconsumptionis

betweenUS$200andUS$400(Table3).

65

Table 3: Results of factor analysis with the loadings of the rotated factors for cognitive variables.

Dimension variable Factor1 Factor2 Factor3 Factor4

Explainedvariance

31.32 15.74 14.85 ‐

Rotatedloadings

BProdInRipAgua 0.82 0.13 0.09 ‐

BScienceTech 0.70 ‐0.05 0.26 ‐

BSoilNeverDegr 0.67 0.22 ‐0.16 ‐

BSoilHealth ‐0.04 0.83 0.19 ‐

BSHchange 0.37 0.7 ‐0.07 ‐

BContrPreci 0.32 ‐0.001 0.67 ‐

Beliefs

BImpoRain 0.11 ‐0.11 ‐0.74 ‐

Explainedvariance

33.18 14.643 10.801 10.459

Rotatedloadings

RPAgrContr 0.65 0.19 ‐0.12 0.42

RPAgrObs 0.85 0.03 0.11 ‐0.09

RPAgrSev 0.76 0.22 0.12 ‐0.10

RPAgrRap 0.72 0.29 0.39 ‐0.15

RPCattleContr 0.59 0.31 0.06 0.43

RPCattleObs 0.27 0.81 0.09 0.20

RPCattleSev 0.21 0.92 0.04 0.06

RPCattleRap 0.12 0.92 0.04 0.01

RPRainObs 0.07 0.03 0.72 ‐0.28

RPRainRap 0.09 0.15 0.63 0.06

RPEroPast ‐0.11 0.06 ‐0.70 ‐0.15

RPCrContr 0.02 0.04 ‐0.19 0.74

Riskperception

RPRainContr ‐0.12 0.11 0.32 0.75

Explainedvariance

33.82 23.39 ‐ ‐

Rotatedloadings

VIncomeNow 0.71 0.41 ‐ ‐

VProdNoPay 0.75 0.25 ‐ ‐

VProdOwnDec 0.73 0.07 ‐ ‐

VRespFutGenr 0.03 0.71 ‐ ‐

Values

VPublicAdmPay 0.005 0.70 ‐ ‐

Themajorityofinterviewees(71%)donot,atpresent,participateinconservationprogramsand

the sample shows a relatively even distribution of farmers in and outside (46%) of risk areas.

66

Farmersinthesample,independentlyfromtheirfarmproductionactivity,implementonaverage

abouthalf(55%)ofthetotalsoilconservationpracticeswelisted.

3.4.3Differentiationoffarmers’groups

Combinedcognitiveandsocioeconomicvariablesclearlydifferentiatethethreegroupsoffarmers

basedontheirconservationefforts(CE ).Agriculturalistsrepresentthemajorityofinterviewees

inthegroups (1)CE and (2)CE (i.e.lowerconservationefforts)(Table4).

Table 4: Distribution of interviewees by farm activity in the different farmers’ conservation- effort groups of the sample.

Farmers’groups(n)

Farmactivity CE(1) CE(2) CE(3)

Agriculture 13 14 8

Dairy‐cattle 3 6 2

Agricultureanddairycattle

3 1 6

Total 19 21 16

Farmerswithinthegroupoflowestconservationefforts (1)CE ownasmallerpercentageofthe

land they cultivate respect to farmers in the other two groups. The bi‐plot of discriminant

analysis6 (Figure 2) identifies three separate groups of farmers distinguished using circles built

with the confidence interval atp‐value lower than 0.05. The first canonical axis explainsmore

than 70% of total variance and allows a clear separation between the group of farmers with

greater conservationefforts (3)CE from thatofgroup (1)CE .Clearly, group (3)CE is strongly

andpositivelyassociatedwithexposureinriskareas,highereducation level,moreparticipation

inprogramsandmore“publicbenefits”(2V )concerns.Thefigurealsoshowsthat inthisgroup,

given thenegativeassociationwith socioeconomicvariables suchas “age” and“own”,we find

younger farmersowninga smallerportionof landwith respect to total land farmed.Similarly,

group (3)CE , incontrastto (1)CE , isalsostronglynegativelyassociatedwithriskperceptionsof

“Agriculturalrisk”andof“extremeprecipitation”.Thedifferencesbetweenthesetwogroupsare

also reflected in their contrasting association with beliefs on topics such as “soil critical

conditions”and“climateextremes”.

6Dataareconsistentforlineardiscriminantanalysissincethenullhypothesesforthehomogeneityofvarianceisaccepted(p=0.4646).

67

Indexes such as “positive thinking” (1B ) beliefs and “pasture risk” (

2RP ) perceptions are only

slightlyimportantwithrespecttothefirstcanonicalaxis.Inthesecondcanonicalaxis, it isonly

cognitiveindexB1whichshowsastrongandpositiveassociationwith (2)CE .Similarly,it isonly

with respect to this axis that the index “private benefits” (1V ) shows a stronger contribution,

beingpositivelyassociatedwith (1)CE andnegativelywithgroup (2)CE .

Figure 2: Discriminant analysis for the three groups of farmers with different soil conservation efforts ( )CE i . Blue triangles correspond to (1)CE ,

yellow squares to (2)CE and green circles to (3)CE . The grey lozenge represents the spatial location of the variable, showing its positive or negative association with farmers groups.

3.4.4Testinghypothesesofdifferencesbetweenfarmers’groups

The clear separation among groups of farmers (defined by their conservation effort) resulting

from the discriminant analysis rejects our null hypothesis supporting that a combination of

socioeconomic and cognitive variables explain significant differences among farmers’

68

conservation efforts. In the following pairwise ANOVA we refine our hypotheses testing to

highlight the differential contribution of cognitive and socioeconomic variables by groups of

farmers.ResultsofANOVAshowkeycontrasting characteristicsof the threegroupsof farmers

(table5).

Table 5: Pair-wise ANOVA comparison of the three farmers’ groups; ( )CE i identifies the Conservation-Effort farmers’ group.

Groupaverages(standarddeviation) p‐values

Variables CE(1) CE(2) CE(3) CE(1)vsCE(2) CE(1)vsCE(3) CE(2)vsCE(3)

PartiProgr 0.11(0.31) 0.14(0.35) 0.69(0.47) 0.48 0.0005 0.001

Activity 2(0.57) 2.24(0.53) 1.75(0.68) 0.23 0.29 0.03

Age 48.53(14.79) 52.57(16.08) 44.81(11.50) 0.39 0.39 0.13

Edu 2.37(0.83) 2(0.83) 2.56(0.89) 0.21 0.44 0.08

MonthCons 4.21(1.18) 4.33(1.71) 4.88(1.02) 0.49 0.11 0.28

Owned 0.70(0.43) 0.83(0.33) 0.70(0.40) 0.33 0.50 0.34

RiskArea 0.21(0.41) 0.48(0.51) 1(0) 0.10 0.0001 0.0002

B1 11.89(4.93) 14.47(4.45) 10.50(4.80) 0.12 0.39 0.02

B2 4.42(1.017) 4(0.94) 3.87(1.08) 0.22 0.17 0.48

B3 9.78(1.81) 10.04(2.17) 8.81(2.28) 0.48 0.22 0.14

RP1 19.84(3.46) 15.90(4.08) 15.93(3.51) 0.003 0.003 0.50

RP2 9.63(6.83) 6.42(4.17) 9.43(4.76) 0.12 0.49 0.07

RP3 17.63(2.31) 16.52(2.69) 17.56(2.09) 0.21 0.49 0.24

RP4 6.47(3.19) 6.81(3.22) 5.56(2.06) 0.49 0.34 0.21

V1 13.42(3.99) 8.38(3.74) 10.25(4.41) 0.0003 0.05 0.22

V2 10.78(3.13) 10.80(2.37) 12.56(1.36) 0.50 0.05 0.01

Farmersingroup (3)CE liveinareasmoreaffectedbyerosionandhavegreaterparticipationin

the soil conservation program than the other two groups showing that technical assistance

programs have been having a significant effect on adoption of soil conservation practices. In

respecttocognitivevariables,higherawarenessofthelimitsoftechnologiesandsoiltoprovide

resilience to erosion together with higher perceptions of i) risks associated to agricultural

activities,andii)benefitsassociatedtosoilconservationandawareness,haveapositiveeffecton

adoption. More specifically, farmers in group (3)CE (who are mainly agriculturalists) show

significantly higher risk perception of the immediacy of pasture impacts on erosion than do

groups (1)CE and (2)CE . Similarly, as opposed to theother twogroups, group (3)CE ismore

concernedwiththelonger‐termbenefitsoferosioncontrol(i.e.forfuturegenerations)aswellas

69

with a perspective that public administration should be paying for removing sediment from

rivers.Ontheotherhand,withrespectto (2)CE , (3)CE farmersshowsignificantlylowervalues

of index B1 (i.e. positive beliefs on the role of technology and soil resilience to reduce the

adverseeffectsofsoilmanagement).

The group with medium conservation effort (2)CE is composed of a significantly larger

proportionoffarmersdedicatedtodairy‐farmingwithalowerparticipationinsoilconservation

programs.Farmersinthisgroupshowasignificantlylowerperceptionoftheriskassociatedwith

humanactivities(i.e.agricultureanddairy‐cattlefarming)thanfarmersintheothertwogroups,

andgenerallylessconcernoverthepublicandprivatebenefitsoferosioncontrol(i.e.V1andV2).

Consistentwiththefindingsofthediscriminantanalysis,riskperceptionofimpactsofagricultural

activitiesonerosionandvaluesregardingshort‐termbenefitsoferosioncontrolhaveaninverse

relationship with conservation efforts. Indeed, farmers with the lowest level of conservation

effort (1)CE showedhigherriskperceptionoftheimpactsofagriculturalactivitiesonerosion(R1

=agriculturalrisk)thanfarmersin (3)CE and (2)CE Farmersingroup (1)CE alsoshowedhigher

concernwithrespecttogroups (2)CE and (3)CE fortheprivateaspectsofsoilconservationand

theimportanceofshorttermbenefitsofagriculturalactivitiesagainstthelongertermbenefitsof

erosioncontrol (V1=“privatebenefits”).Deeperanalysisof thedataontenureandproduction

activities will help to illustrate this finding. In group (1)CE we findmainly horticulturalists as

opposed to the other two groups (where horticulturalists and dairy‐farmers are more evenly

represented). A comparison of the average farmland owned by these two types of producers

showsthathorticulturalistsownsmallerplots( )0.005Kruskal Wallisp!

< inwhichtheopportunitycost

ofsoilconservationmightbeperceivedasbeinghigher.Small‐scalefarmers,dealingwithshort‐

term production cycles, such as is the case of horticulture in the area, however, might show

awarenessofsoildegradationbutfeelthatprivatebenefitsaremoreimportantintheshortrun.

To furtherexplore thesehypotheseswe rananon‐parametric test to comparemeanvaluesof

cognitive indexes accounting for farmers’ groups ( )CE i and also accounting for size of farm

owned. We thus categorized the variable “total farmland” available to producers into two

separate classes; namely, those owning less than 10 ha ( 16n = ) and the others ( 40n = ). For

smaller landowners, the groupwith lower conservation efforts (1)CE has a significantly higher

70

score for beliefs about “critical soil conditions” ( 0.05)Kruskal Wallisp!

< and “private benefits”

compared to the other two groups ( 0.05)Kruskal Wallisp!

< . Finally, the producers in group (3)CE

showgreaterconcernsabout“publicbenefits”(2V )aspectsthantheothertwogroupsoffarmers.

3.5Discussion

Theresultsofouranalysisrejectsthehypothesisthatfarmers’voluntarysoilconservationefforts

arenotinfluencedbytheinteractionofcognitiveandsocioeconomicvariables,keepingwiththe

findings of previous research in the region (Foster, 1994). Factor analysis allowed the

identificationofthemost importantcognitiveaspectsthat influencesoilconservationdecisions

andtheconstructionofconsistentindexes.Themultivariatediscriminantanalysisindicatesthat

the group of farmers with lower conservation effort is positively correlated with cognitive

variables. This contrasts with the group showing higher conservation effort (i.e. group (3)CE ),

where there is mainly a positive correlation with socioeconomic variables and exposure to

erosionrisk.Thissuggeststhat, togetherwiththepromoteddirectpayments,acomplexsetof

factorsneedtobeconsideredindesigningconservationprogramsaimedatpromotingadoption

ofappropriatesoilmanagementpractices.

Our results indicate that complex interactions among cognitive and socioeconomic dimensions

influence significantly farmers’decisionson soil conservation.However, comparisonwithother

studies in the Central American region can only be limited to socioeconomic and some

institutionalvariables(e.g.presenceofsoilconservationprograms)whilecognitivevariableshave

been overlooked. More specifically, from an economic perspective, farmers with lower

conservation effort are also those cultivating smaller farmlands, which might entail the

perceptionofhigheropportunitycostsintheimplementationofsoilconservationpractices.This

is supportedby the resultsof theeconomicanalysisof soil conservationpractices in theclose

Tierrablanca watershed in Costa Rica (Cuesta, 1994) where results indicate that conservation

practiceshavenegativereturnsalsoduetotheproportionofproductivefarmlanddedicatedto

conservation (i.e. small farms having higher opportunity cost). Our results also support the

finding of Cuesta (1994) and Vasquez and Santamaria (1994) by suggesting that institutional

programspromoting technical assistance should be strengthenedgiven their positiveeffect on

adoptionofsoilconservationpractices.Ourfindingsshow,however,thatcognitivevariablesalso

71

influencetheextenttowhichindividualfarmersdecidetoimplementsoilconservationpractices.

Farmers’ evaluation of the benefits of soil conservation for agricultural production trades off

against the costs they attach to erosion control. Their judgment on the costs of soil erosion

controlandonitsbenefitsappearstobeconditionedbytheirspecificcontext,suchastheextent

and tenure regimeof the landcultivated. Smallerholdingsareassociatedwithhorticulturalists

thathavelesslandavailablefortheimplementationofconservationpracticesand,moreover,are

moreconcernedwithshort‐termincomeneeds,althoughtheyscorehigheronawarenessofrisks

andcausesofsoilerosioninthearea.

Farmersinoursamplewith lowerconservationeffortsshowasignificantlyhigherperceptionof

the risk of agricultural activities. This finding contrastswith those ofWeber et al. (2001),who

reported that actual exposure and awareness increase proactive behaviour. Additionally, the

“availabilityheuristic”paradigm(TversyandKahneman,1974)mayalsobeplayinganimportant

role in farmers’ soil conservation behaviour. Indeed, the inverse relationship between risk

perceptionofthe impactsofagriculturalactivitiesonerosionandconservationeffortssuggests

that farmersmayunderestimatetheconsequencesoftheiragriculturalactivitiesonerosion. In

other words, daily experience with erosion might give the illusion of control and/or smaller

losses. Moreover, as discussed Bewket (2007), while awareness of risk and its causes can

influence farmers’ intentions to implement soil conservation practices, its actual adoption

dependsonsocioeconomicandinstitutionalaspects.

Themosteducatedfarmersdonotshowstrongbeliefsregardingthesoil’scapacitytorecoveror

theabilityofscienceandtechnologytoovercomeerosionproblems(2B )andtheyrevealhigher

conservationefforts. These findings contrast thoseofAbreu (1994)whom foundno significant

effectofeducationonadoptionleveloffarmersintheclosewatershedofTierrablanca.Educated

farmers in our sample show awareness of the importance of implementing soil conservation

practices, understanding that soil productivity can be degraded if appropriate actions are not

initiated.Thesefindingsconfirmthoseofotherauthors, inkeepingthateducationisrelatedto

knowledgeofconsequencesofsoilmanagementpracticesandofalternativesolutions,which in

turninfluencesbehaviour(Carlsonetal.,1977;ErvinandErvin,1982;Traoreetal.,1998;Mbaga‐

SemgalaweandFolmer,2000;Lambertetal., 2007). In linewiththissameargument, technical

72

assistance programs improving farmers’ understanding of erosion causes and alternatives

availablecanimprovetheirchancesofadoption(Traoreetal.,1998).

3.6Conclusions

This article provides new perspectives on the relationship between the perception of climate

changeriskposedtofarmsoilsandtheadoptionofconservationpractices.Themodelweused

enriches our understanding of the complex interactions among socioeconomic and cognitive

variablesbyshowinghowtheseinteractionsdifferentiateamongspecificgroupsoffarmertypes.

Farmers showa lowawarenessof the risksposedby climate‐change‐relatedextremes to their

farmsoil, although they showhighconcernwith the impactsofhumanactivities.Additionally,

our results suggest that soil conservation programs should strategically consider promoting

improvement of understanding of the causes, the on‐site and off‐site consequences, and the

limitationofsoilsandtechnologicalsolutionstoprovideresiliencetoerosion.Thismightimprove

understandingontheurgencytotakeactiongiventhepotentialintensificationoferosion‐causing

extremeeventsinvulnerablewatershedsand increasetheprobabilitiesof largerself‐motivated

adoptionofsoilconservationpractices.

3.7References

Abreu,H.M.,1994.AdoptionofsoilconservationinTierraBlanca,CostaRica.In:Lutz,E.,Pagiola,S.,Reiche,C.(Eds.),Economicandinstitutionalanalysisofthesoilconservationprojects inCentralAmericaandtheCaribbean.WorldBankDevelopmentpaperNo.8,WashingtonD.C.,USA,207pp.

Aguilar,E.,Peterson,T.C.,RamırezObando,P.,Frutos,R.,Retana,J.A.,Solera,M.,Soley,J.,GonzalezGarcıa,I.,Araujo,R.M.,RosaSantos,A.,Valle,V.E.,Brunet,M.,Aguilar,L.,Alvarez,L.,Bautista,M.,Castañonp,C., Herrera, L., Ruano, E., Sinay, J.J., Sanchez, E., Hernandez Oviedo, G.I., Obed, F., Salgado, J.E.,Vazquez, J.L., Baca, M., Gutierrez, M., Centella, C., Espinosa, J., Martınez, D., Olmedo, B., OjedaEspinoza,C.E.,Nuñez,R.,Haylock,M.,Benavides,H.,Mayorga,R.,2005.Changes inprecipitationandtemperature extremes in Central America and northern South America, 1961‐2003. Journal ofGeophysicalResearch,110,D23107.

Ajzen,I.,1991.TheTheoryofPlannedBehaviour.OrganizationalBehaviourandHumanDecisionProcesses50,179‐211.

Bayard, B., Jolly, C., 2007. Environmental behaviour structure and socio‐economic conditions of hillsidefarmers:amultiple‐groupstructuralequationmodelingapproach.EcologicalEconomics,62(3/4),433‐440.

Bewket,W., 2007. Soil andwater conservation interventionwith conventional technologies in northwesternhighlandsofEthiopia:acceptanceandadoptionbyfarmers.LandUsePolicy24,404‐416.

Bruijnzeel, L.A., 2004.Hydrological functionsof tropical forests:not seeing thesoil for the tres?. Agriculture,EcosystemsandEnvironment104,185‐228.

73

Comisión Asesora sobre Degradación de tierras (CADETI), 2002. Segundo informe nacional para elcumplimientodelaUNCCDenCostaRica.SanJosé,CostaRica,56pp.

Carlson, J.E., McLeod, M., Lassey,W.R., Dillman, D.A., 1977. The farmer, absentee landowner and erosion:factorsinfluencingtheuseofcontrolpractices.IdahoWaterResourcesInstitute,Moscow,Idaho.

Cuesta,M.D.,1994.EconomicanalysisofsoilconservationprojectsinCostaRica.In:Lutz,E.,Pagiola,S.,Reiche,C. (Eds.),Economicand institutionalanalysisof thesoil conservationprojects inCentralAmericaandtheCaribbean.WorldBankDevelopmentpaperNo.8,WashingtonD.C.,USA,207pp.

D’emden,F.H., Llewellyn, R.S.,Burton,M.P.,2008.Factors influencing theadoptionofconservation tillage inAustralian cropping regions. The Australian Journal ofAgricultural and Resource Economics 52, 169‐182.

Dale, V.H., Polansky, S., 2007. Measures of the effects of agricultural practices on ecosystem services.EcologicalEconomics64,286‐296

Ervin, C.A., Ervin,D.E., 1982. Factors affecting theuseof soil conservationpractices: hypotheses, evidences,andpolicyimplications.LandEconomics58(3),277‐292.

Featherstone, A.M., Goodwin, B.K., 1993. Factors influencing a farmer’s decision to invest in long‐termconservation.LandEconomics69(1),67‐81.

Forster,N.R.,1994.TenuresystemsandresourceconservationinCentralAmerica,Mexico,HaitiandDominicanRepublic. In: Lutz, E., Pagiola, S., Reiche, C. (Eds.), Economic and institutional analysis of the soilconservationprojects in Central America and the Caribbean.World BankDevelopment paperNo. 8,WashingtonD.C.,USA,207pp.

Gould, B.W.,Saupe,W.E., Klemme,R.M.,1989. Conservation tillage: the roleof farm characteristicsand theperceptionofsoilerosion.LandEconomics65(2),167‐182.

Grasmuck,D.,Scholz,R.W.,2005.Riskperceptionofheavymetalsoilcontaminationbyhigh‐exposedandlow‐exposedinhabitants:theroleofknowledgeandemotionalconcerns.RiskAnalysis25(3),611‐622.

Grothmann,T.,Patt,A.,2005.Adaptivecapacityandhumancognition:theprocessofindividualadaptationtoclimatechange.GlobalEnvironmentalChange15,199‐213.

GuoZ.,XiaoX.,LiD.2000.Anassessmentofecosystemservices:waterflowregulationandhydroelectricpowerproduction.EcologicalApplications10(3),925‐936.

Illukpitiya, P., Gopalakrishnan, C., 2004. Decision‐making in soil conservation: application of a behaviouralmodeltopotatofarmersinSriLanka.LandUsePolicy21,321‐331.

InstitutoCostarricensedeElectricidad(ICE),2000.PlandeManejodelaCuencadelRíoReventazón.SanJosé,CostaRica.

Instituto Nacional De Estadística y Censo Centroamericano de Población (INEC), 2002. Estimaciones yproyeccionesdepoblación1970‐2050.SanJosé,CostaRica,112pp.

Innes, R., Ardila, S., 1994. Agricultural insurance and soil depletion in a simple dynamic model. AmericanJournalofAgriculturalEconomics76(3),371‐384.

Kinnel,J.,Lazo, J.K.,Epp,D.J.,Fisher,A.,Shortle, J.S.,2002.Perceptionsandvaluesforpreventingecosystemchange:PennsylvaniaDuckHuntersandthePrairiePotholeRegion.LandEconomics78(2),228‐244.

Lambert, D.M., Sullivan, P., Claassen, R., Foreman, L., 2007. Profiles of US farm household adoptingconservation‐compatiblepractices.LandUsePolicy,24,72‐88.

Lee,H.F.,Zhang,D.D.,2005.Perceivingland‐degradationactivitiesfromthelayperspectiveinnorthernChina.EnvironmentalManagement36(5),711‐725.

Leyva,J.C.,Martinez,J.A.F.,Gonzalez‐Roa,M.C.,2007.Analysisoftheadoptionofsoilconservationpracticesinolivegroves:thecaseofmountainousareasinsouthernSpain.SpanishJournalofAgriculturalresearch5(3),249‐258.

Lutz,E.,Pagiola,S.,Reiche,C.,1994a.Thecostsandbenefitsofsoilconservation:thefarmers’viewpoint.TheWorldBankResearcherObserver9,273–295.

74

Lutz,E.,Pagiola,S., Reiche,C.,1994b.Economicand institutionalanalysisof thesoil conservationprojects inCentralAmericaandtheCaribbean.WorldBankDevelopmentpaperNo.8,WashingtonD.C.,207pp.

Lynne,G.D.,Shonkwiler, J.S.,Rola,L.R.,1988.Attitudesandfarmerconservationbehaviour.AmericanJournalofAgriculturalEconomics70(1),12‐19.

Magrin,G.,GayGarcía,C.,CruzChoque,D.,Giménez,J.C.,Moreno,A.R.,Nagy,G.J.,Nobre,C.,Villamizar,A.,2007. Latin America. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution ofWorkingGroupIItotheFourthAssessmentReportoftheIntergovernmentalPanelonClimateChange,Parry,M.L.,Canziani,O.F.,Palutikof,J.P.,vanderLinden,P.J.,Hanson,C.E.,Eds.,CambridgeUniversityPress,Cambridge,UK,581‐615.

Marchamalo, M., 2004. “Ordenación del territorio para la producción de servicios ambientales hídricos.aplicaciónalacuencadelRíoBirrís(CostaRica).PhDThesis,ForestryDepartmentofthePolytechniqueUniversityofMadrid,Spain,409pp.

Marchamalo,M.,Romero,C.,2007.Participatorydecision‐makinginlanduseplanning:anapplicationinCostaRica.EcologicalEconomics63(4),740‐748.

Mbaga‐Semgalawe, Z., Folmer, H., 2000. Household adoption behaviour of improved soil conservation: thecaseoftheNorthPareandWesternUsumbaraMountainsofTanzania.LandUsePolicy17,321‐336.

Mcdaniels, T.L., Axelrod, L.J., Canavagh, N.S., Slovic, P., 1997. Perception of ecological risk to waterenvironment.RiskAnalysis17(3),341‐352.

McDaniels, T.L., Axelrod, L.J., Slovic, P., 1996. Perceived ecological risks of global change: a psychometriccomparisonofcausesandconsequences.GlobalEnvironmentalChange6(2),159‐171.

Ministerio de Medio Ambiente y Energía (MINAE), 2002. GEO Costa Rica: una perspectiva sobre el medioambiente.ObservatoriodeDesarrollodelaUniversidaddeCostaRica,SanJose162pp.

O’Connor,R.E.,Bord,R.J.,Fischer,A.,1999.Riskperceptions,generalenvironmentalbeliefs,andwillingnesstoaddressclimatechange.RiskAnalysis19(3).461‐471.

Pagiola,S.,2008.PaymentsforenvironmentalservicesinCostaRica.EcologicalEconomics65(4),712‐724.Parry, M.L., Canziani, O.F., Palutikof, J.P., Van der Linden, P.J., Hanson, C.E., 2007. Climate Change 2007:

Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth AssessmentReport of the Intergovernmental Panel on Climate Change, Intergovernmental Panel on ClimateChange(IPCC),CambridgeUniversityPress,Cambridge,UK,976pp.

Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., McNair, M., Crist, S., Shpritz, L., Fitton, L.,Saffouri, R., Blair, R., 1995. Environmental and economic costs of soil erosion and conservationbenefits.Science267(5201),1117‐1123.

Ramirez,L.C.,2008.Caracterizaciónfísica,socioeconómicaeindicadoresdesostenibilidadenlaCuencamediadelRioReventado.MScThesis,CiudadUniversitariaRodrigoFacio,SanJosé,CostaRica,192pp.

Sanchez‐Azofeifa,G.A.,Harriss,R.C.,Storrier,A.L.,DeCamino‐Beck,T.,2002.WaterresourcesandregionallandcoverchangeinCostaRica:impactsandeconomics.WaterResourcesDevelopment18(3),409‐424.

Scholz,R.W.,n.d.EnvironmentalLiteracy inScienceandSociety:FromKnowledgetoDecisions”byRolandW.Scholz will be available in early 2011 at Cambridge University Press (Hardback 978‐0‐521‐19271‐2,Paperback978‐0521‐18333‐8).

Siegrist, M., Gutscher, H., 2006. Flooding risks: a comparison of lay people’s perceptions and expertsassessmentinSwitzerland.RiskAnalysis26(4),971‐979.

Slovic,P.,1987.PerceptionofRisk.Science236(4799),280‐285.SoilandWaterConservationSociety(SWCS),2003.Soilerosionandrunofffromcropland.ReportfromtheUSA,

SoilandWaterConservationSociety63pp.Soule,M.J.,Tegene,A.,Wiebe,K.D.,2000.Landtenureandtheadoptionofconservationpractices.American

JournalofAgriculturalEconomics82(4),993‐1005.Southgate, D., Macke, R., 1989. The downstream benefits of soil conservation in third world hydroelectric

watersheds.LandEconomics65(1),38‐48.SPSSforWindows,Rel.11.01.1Chicago:SPSSInc.

75

Swinton, S.M., Lupi, F., Robertson, G.P.,Hamilton, S.K., 2007. Ecosystem services and agriculture: cultivatingagriculturalecosystemsfordiversebenefits.EcologicalEconomics64,245‐252.

Tejwani, K.G., 1993.Watermanagement issues: Population, Agriculture and Forests ‐ a focus on watershedmanagement. In: Bonell, M., Hufschmidt, M.M., Gladwell, J.S. (Eds.), Hydrology and WaterManagementintheHumidTropics.UNESCO,Paris,pp496‐525.

Traore, N., Landry, R., Amara, N., 1998. On‐farm adoption of conservation practices: the role of farm andfarmercharacteristics,perceptions,andhealthhazards.LandEconomics74(1),114‐127.

Tversky,A.,Kahneman,D.,1974.Judgmentunderuncertainty:heuristicsandbiases.Science185(4157),1124‐1131.

VandeBergh,J.C.J.M.,Ferrer‐i‐Carbonell,A.,Munda,G.,2000.Alternativemodelsofindividualbehaviourandimplicationsforenvironmentalpolicy.EcologicalEconomics32,43‐61.

Vasquez,T.,Santamaria, J.,1994.Economicandinstitutionalanalysisofsoilconservationatthefarm level inCoclé,Panama.In:Lutz,E.,Pagiola,S.,Reiche,C.(Eds.),Economicandinstitutionalanalysisofthesoilconservationprojects in Central America and the Caribbean.World BankDevelopment paperNo. 8,WashingtonD.C.,USA,207pp.

Vignola, R., Otarola, M., Calvo, G., 2010. Defining ecosystem‐based adaptation strategies for hydropowerproduction:stakeholders’participationindevelopingandevaluatingalternativelandusescenariosandthe strategies to achieve desired goals. In Martinez, C., Locatelli, B., Imbach, P., Vignola, R., (Eds),2009.Proceedings of the international Seminar on “Adaptation to Climate Change: the role ofecosystemservices”,CATIE,Turrialba,CostaRica.InPress.

Wagner,K.,2007.Mentalmodelsofflashfloodsandlandslides.RiskAnalysis27(3),671‐682.Weber,O.,Scholz,R.W.,Bühlmann,R.,Grasmück,D.,2001.Riskperceptionofheavymetalsoilcontamination

andattitudestowardsdescontaminationstrategies.RiskAnalysis21,(5).967‐977.Wischmeier,W.H.,Smith,D.D.,1978.PredictingRainfallErosionLosses.AGuidetoConservationPlanning.US

DepartmentofAgricultureHandbookNo.537.USGovernmentPrintingOffice,WashingtonD.C.,USA.Wossink, A., Swinton, S.M., 2007. Jointness in production and farmers’ willingness to supply non‐marketed

ecosystemservices.EcologicalEconomics,64,297‐304.Wunder,S.,Engel,S., Pagiola,S.,2008. Takingstock:acomparativeanalysisofpayments for environmental

servicesprogramsindevelopedanddevelopingcountries.EcologicalEconomics65(4),834‐852.

77

4.1 Negotiation analysis for mechanisms to deliver ecosystem services: a value-focused bargaining structure to achieve joint gains in the provision of Soil Regulation Services

Authors:RaffaeleVignola,TimL.McDaniels,andRolandW.Scholz

Paperstatus:undersecondroundrevisioninEcologicalEconomics.

Abstract

Directpaymentschemeshavebeendiscussedintheliteratureaseconomicallyefficientmeansto

createandmaintainthe institutionalandpracticallinkagestoreducedegradationofecosystem

services (ES). However, evidence suggests that the nature and structure of institutional

mechanisms is fundamental to the efficacy and efficiency of outcomes, yet has received little

attention.

Linkages between ES producers and consumers require bargaining about the characteristics of

mechanismsforcooperationatagivenlocation.Inthispaper,weintroducenegotiationanalysis

informedbybothstakeholders'value‐basedinformationandtechnicalknowledgeasameansof

structuring negotiation about mechanisms for maintaining ES. We use a case study in a

watershedwhereupstreamfarmersanddownstreamhydropowermightbenefitfromthedesign

of a mechanism to foster the provision of soil regulation services. We identify the values of

producersandconsumerstounderstandtheirviewson importantmeansandends indecisions

about ES provisions. We complement this value‐based information, with literature on

alternatives forsoil conservationprograms.We, thus,createanegotiationtemplateto identify

bargainingopportunitiesandthusthepotentialforbetteroutcomes.Ourresultsshowthatthere

is convergence of preferences on the important aspects of themechanisms between the two

stakeholders.Thisprovidesopportunitiesforjointgainsintheprovisionofsoilregulationservices

thatcanprovidemoreresiliencetoclimatechangeinCostaRica,whichisexpectedtobringmore

erosionduetomoreintenserainfall.

Keywords:Negotiationanalysis,decisionanalysis,ecosystemservices,soilconservation

4.1Introduction

TheMillenniumEcosystemAssessment documents themassive growth in humandemands for

services fromecosystemsover the last fourdecadesand revealsthatover two thirdsofglobal

78

ecosystemservices(ES)systemsareindecline(MEA,2005).ItprovidesatypologyofdiverseES

produced within linked human and natural systems7. ES involving water resources are often

produced or regulated at upstream locations and consumeddownstream (Kosoy et al., 2007).

This is thecaseofsoil regulationservices (SRS)whoseprovision is influencedbyactorssuchas

upstreamfarmersaffectingthequantityofsedimentsinriversanddownstreambeneficiariesof

SRSsuchashydropowerdams(SouthgateandMacke,1989).Oftenthesegroupsarespatiallyand

administratively separated,donothaveahistoryof interactions,and lackanunderstandingof

thepotentialjointgainsifupstreamproducershadtheinterestsofdownstreamconsumersmore

directlyinmind,andhadresourcestoaddressthoseinterests(Isonetal.,2007).

Themechanismmostwidelyemployedtolinkproducersandconsumersofecosystemservicesis

direct (or indirect) Payment for Ecosystem Services (PES) from direct beneficiaries (or

governments) to providers. Researchers have provided typologies of PES compared to other

policy instruments (Engel et al., 2008), and considered ways to improve economic (Ferraro,

2008),environmental(Wunscheretal.,2008)andsocial(Pagiolaetal.,2005)performanceofPES

schemes.TheunderliningreasoningofPESschemesputsemphasisonthepotentialforefficiency

gains bymaking externalities internal to the decisions of ES providers. Yetmany social factors

otherthanopportunitycostsinfluencethedecisionsofactorstoalterlanduses,asconfirmedby

acomparativestudyintheCentralAmericanregionwhichshowedthatproducersgenerallyfailed

tomeettheiropportunitycostsineffortstoprovideES(Kosoyetal.,2007).

Henceabroaderformofbargaining,inwhichmultipleobjectivesareconsideredonbothsidesof

ecosystem service production and consumption, built on an understanding of how upstream

producerscouldbemotivatedtoprovideservicesfordownstreambeneficiaries,wouldbehighly

relevant in designing mechanisms for ES provision. Previous experience in cooperative

managementofnaturalresourceshasshownthatformalandinformalinstitutionalarrangements

arefundamentaltooutcomes(Dietzetal.,2003).Yethowmechanismsaretobebuiltinorderto

foster ES provision has received relatively little attention (Wilson and Horwath (2002) and de

Groot and Herman (2009), are exceptions). This paper attempts to address that gap through

7 These include:provisioning services suchas food,water, timber, and fiber; regulating services thataffect climate, floods,

disease,wastes,andwaterquality;culturalservicesthatproviderecreational,aesthetic,andspiritualbenefits;andsupportingservicessuchassoilformation,photosynthesis,andnutrientcycling(MEA,2005:9).

79

attentiontobargainingfordesignofmechanismsforprovisionofESthataremutuallyagreeable

tothe interestedparties.Thestandardnormativeapproachtoconflictandnegotiation isgame

theory, which seeks normative equilibrium solutions given known alternatives, constrained

objectives, and known payoffs8. Here we adopt a more behaviourally‐oriented, prescriptive

approachtoanalysisofbargainingcontexts,withanemphasisondecisionstructureandprocess,

referredtoasnegotiationanalysis(Young,1991;Sebenius,1992;Raiffaetal.,2002).

CostaRicahasexperiencewithPESthroughitsrecentlyestablishedsystemofpayinglandholders

formaintainingorregeneratingforestcover(Pagiola,2008).TheperformanceofPESschemesin

CostaRicahasbeenthesubjectofmuchdiscussion,reviewandevaluation(Mirandaetal.,2007;

Sanchez‐Azofeifaetal.,2007;Wünscheretal.,2008).Inthispaper,weconsideranothercontext

for ecosystem service provision in Costa Rica, involving farmers who are horticulturalists and

dairy‐cattleproducersonsteep,cool,erosion‐pronevolcanicslopesintheBirriswatershed,high

abovereservoirsthatareimportantfornationalhydropowerproduction.Weusethecontextof

potential SRS provision through soil erosion control to develop a bargaining framework for a

mechanism to foster ES provision, in which the downstream power producers could provide

incentives or other kinds of support. Improving soil erosion controlwould benefit two electric

utilitiesbyreducingtheamountofsediment in rivers, (theAdministrativeBoardoftheElectric

UtilityofthecityofCartago(JASEC)andtheCostaRicanInstituteforElectricity(ICE)),aswellas

providingon‐sitebenefitsforthefarmersengagedinserviceprovision(Pimenteletal.,1995;Guo

etal.,2000;WossinkandSwinton,2007).

This paper has two related objectives. One objective is to demonstrate the relevance of

negotiation analysis for complex bargaining contexts such as developing mechanisms for ES

provision. A second objective is to provide a structure for the negotiation space among

stakeholders.Webuildatemplatetoguidebargaininginthedesignofmechanismstoaccomplish

linkages among producers and consumers. ES provision in the Birris watershed is used as an

8Forarichdiscussiononthelinkagesanddifferencesbetweengametheoryandnegotiationanalysis,seeSebenius(1992).Alivelyexchangebetweennormativescholarsofgametheoryandmorebehaviourally‐orientedBayesianscholarsoccurredin

thepagesofManagementSciencefrom1982to1983(see,forexample,KadaneandLarkey, (1982,1983);Harsanyi,(1982),RothandSchoumaker(1983),andRothkopf(1983).Oneauthorsharpenedthedebatetoaquestionposedinthetitleofhis

paper: "OnChoosingBetweenRev. BayesandProf. vonNeumann" (Kahan (1983), as cited inSebenius,1992).Raiffaatal.(2002)alsoprovidesacomparisonofgametheoryandnegotiationanalysis.

80

example. The next section provides an introduction to negotiation analysis, and the particular

characteristics of ES provision thatmake the approach especially relevant. Section 3 provides

moredetailsoftheBirrisecosystemservicescontextrelevantfornegotiationanalysis.Section4

discussesmethods used to gather information for this research. Section 5 includes results, in

termsofanegotiationtemplateanditsevaluationbyfarmersandtheelectriccompany.Section

6providesconclusions.

4.2NegotiationAnalysisandItsRelevanceforES

Collaborative planning for complex human‐natural systems necessarily relies on negotiation

(InnesandBooher,1999,Fisheretal.,2009);yethowtostructurenegotiationsinsuchcontexts

meritsgreaterattention.Negotiationanalysiscouldpotentiallybeavaluabletoolforbuildingthe

basis foragreementamongconflicting intereststhatcharacterizethesecontexts (deGrootand

Hermans, 2009). Negotiation analysis might be referred to as both “game theory minus” and

“decisionanalysisplus”(Sebenius,1992).

“Game theory minus” indicates that while it deals with the same broad kinds of issues,

negotiation analysis does not entail the examination of well‐defined gameswith known rules,

playersandpayoffs,inasearchfornormativeequilibriumsolutions,asingametheory.Ratherit

is asymmetrically prescriptive/descriptive in orientation, providing prescriptive guidance on

negotiationdecisionprocess foroneactor,basedondescriptionsofexpectedbehaviourbythe

other party9. Negotiation analysis has a Bayesian (subjectivist) perspective on uncertainty and

information,andadoptsboundedrationalityratherthannormativeoptimizationastheguiding

assumption regarding behaviour of participants. It recognizes the benefits of full and open

informationexchangeinnegotiation10.

“Decisionanalysisplus”indicatesthatnegotiationanalysisinvolvessituationsinwhichagreement

among multiple parties is required, providing the potential for joint gain, which are more

demandingcontextsthananalysis forasingledecisionmaker. Itdrawsonthedecisionanalytic

process(Hammondetal.,1999)intermsofaseriesofiterativestepsinvolvingvalueperspectives

9 In this paper, we seek to provide support and analysis for both sides in potential negotiations, so its orientation isprescriptive,builtondescriptiveunderstandingforbothsides.10Raiffaetal.2007referstothebenefitsof“full,open,truthfulexchange”(FOTE)asanidealinreachingbetteragreementswithinnegotiationanalysis.

81

andtechnicalinformation,withafocusonvaluesofthepartiestothenegotiationasabasisfor

constructing and comparing alternatives. Raiffa (1982) first explicitly considered how the

concepts andmethods of decision analysis for a single decisionmaker could be broadened to

serveasthebasisforbothanartandscienceofnegotiation.Intheearly1990s,asurveyarticle

(Sebenius, 1992) and edited book (Young, 1991) delineated negotiation analysis as a new

perspective on conflict and bargaining, and provided concepts and applied examples to guide

theseefforts.ArecentbookbyRaiffaandcolleagues(2002)furtherdevelopsnegotiationanalytic

concepts and practice. Drawing on these sources, key aspects of negotiation analysis that are

particularlyimportantforESincludethefollowingpoints.

Creatingorchangingthenegotiation.ESprovisionisacontextinwhichthepotentialparties(i.e.

producersandconsumers)toaprospectivenegotiationmaybespatiallydistant,ofdifferentclass

or ethnic backgrounds, and operate in entirely different institutional or governance contexts.

Theymayhavelittleornoknowledgeofthepotentialgainsthatcouldbeachievedforconsumers

ifproducersmanagedESprovisiondifferently.Hencethepreceptsofagamewithknownparties,

known rules and knownpayoffs forwell‐defined alternatives are not relevant. In ES provision,

there isaneed tobeginby creating thecontext foranegotiation,orperhapsalteringexisting

arrangements,ratherthanlearningthroughrepeated,well‐definedgames.

Afocusonthevaluesofthenegotiatingparties.Thevaluesofthepartiesarethereasonwhy

theymaywishtonegotiateandachievejointgains.ValuesforESprovisionmaybecomplexand

involvedifferentconcernsfordifferentgroups(Isonetal.,2007).Forexample,theworkbyKosoy

andcolleagues(2007)showsthatmultipleobjectivesnecessarilymatter forESprovision.While

almost any approach to decision‐making or negotiation calls for attention to values, decision

analyticapproacheshavewell‐developedandeffectivemethodsto characterizevalues.Keeney

(1992)discussesvalue‐focusedthinkingasaframeworkforcharacterizingvaluesforanydecision

context,includingmulti‐partynegotiation,(seeSection4).

Creating attractive alternatives that expand the benefits, and are more likely to facilitate

agreement.Negotiationanalysisbuiltonvalue‐focused thinkingattempts to createnew,more

attractivealternativesthatarebothmore likelytobe implemented,andmore likelytoaddress

thefullrangeofobjectivesamongtheparties(Keeney,1996).Inthismanner,thepartieshavean

opportunitytoexpandthepotentialbenefitsofagreement,ratherthanonlyfocusingonclaiming

82

benefits11.Reachingagreementsbasedonrelevantstakeholders’valuescansupportthedesign

ofmoresocially‐stablemanagementstrategies(e.g.buildingtrustandmutuallearning)thatare

importantforlonger‐termadaptivesolutionsneededinESprotection(InnesandBooher,1999).

How the alternatives are to be constructed is a particularly significant question involving

technicalunderstanding,knowledgeofrelevantvalues,andcreativityinlinkingthetwo.Keeney

stressestheimportanceofvaluesasaguidetodesignofalternatives,anddifferentiatesamong

valuesfortheoutcomes(theultimateendsofinterest,suchasESprovision,costsandsoforth)

versusvaluesforthedesignoftheprocess.

Attention to the alternatives to a negotiated agreement.Anegotiator’s best alternative to a

negotiated agreement (BATNA) is widely seen as an important indicator of the value of an

alternative and thus negotiation strategy. In the case of ES provision, the BATNAmay be the

statusquo.Alternatively,theremayberecognitionthatcompetitionforESisincreasinginmany

locations,andglobalchangeprocesses,particularlyclimatechange,maywell limittheextentto

whichthestatusquoofESprovisioncancontinue.Understandingthatthestatusquomaynotbe

viableoverthemediumtolongtermprovidesmoreincentivetoreachanegotiatedagreement

forESprovision.

4.3ESprovisionintheBirrisWatershed

TheBirriswatersheddescendsfromtheslopesofIrazuVolcanoabovethecentralvalleyofCosta

RicainCartagoprovince.Thewatershedischaracterizedbyrichanddeepsoils(i.e.andosols)that

have attractedmany farmers to cultivate vegetables on small tomedium sized plots for local

markets. Over 300 producerswork in thewatershed, often on very steep slopes, in locations

whereannualprecipitationis2325mm(Marchamalo,2004).Extremeprecipitationeventshave

increasedoverthelastfortyyears(Aguilaretal.,2005)andareexpectedtocontinuetoincrease

asaresultofclimatechange(Magrinetal.,2007).

AresearchprojectwithmanyregionalpartnershasinvestigatedseveralaspectsofSRSprovision

in the Birriswatershed, and its links to hydroelectric facilities downstream. These facilities are

greatlyaffectedbysiltation,which isexpected toworsen in the futureunder changingclimate

11 Sebenius (1992)providesadiscussionof the tensionbetweencreatingandcreatingvalue innegotiationanalysis, andidentifieswhathetermsthe“negotiator’sdilemma”,incontrasttothePrisoner’sdilemma.

83

conditions.BothJASECandICEoperateelectricalgenerationfacilitiesdownstreamfromtheBirris

watershed,forwhichcostlydredgingandflushingofsilt,orreplacementofenergypoweredby

diesel generators, or both, are needed to address power losses due to siltation in reservoirs.

Interviewsindicatedthat,in2006,morethan$fourmillionUSinforegonepowerproductionand

otherdirectcostswerespentbytheutilitiestoflushsedimentsfromoftheirreservoirs.

ThesituationintheBirriswatershedillustratesseveralofthepointsmadeintheprevioussection

aboutthenatureofESprovisionandwhynegotiationanalysis is relevant.Theutilitieshaveno

direct experience in building a large scale program to provide incentives or other support to

farmers for soil conservation practices. Neither ES providers nor beneficiaries in the Birris

watershed have a clear understanding of the need to bargain or the potential benefits of

bargaining. They operate at different scales of governance, and different spatial scales, and

require new institutionalmechanisms in order to address joint interests.While farmers in the

watershedunderstanderosionprocesses,theyhavelimitedawarenessoftheeffectsoferosion

ondownstreamhydropowerfacilitiesandofhowtheeffectsofclimatechange(inmakingrainfall

moreintense)canaffecttheirownfarmsoils.

NationallawrequireselectriccompaniestopayataxtotheMinistryofEnvironmentMINAETfor

wateruse.Apartofthesepaymentscanbedestinedtofundwatershedmanagementactivities

(Ponce,2006).Thetwodownstreamhydropowercompanies,JASECandICEhaveinitiatedsome

small scaleprograms to i) increaseawareness, ii) encouragevoluntaryerosioncontrol, and, iii)

provide some learning opportunities for soil conservation practices. Yet the companies have

limitedunderstandingof farmers’objectives,andtheconstraintstheyface. Inaddition,neither

farmers nor the electric companies havemuch experiencewith the costs and structure of soil

conservationmechanisms,ortheirbenefitsintermsofactualchangesintheratesoferosion.

Althoughecosystemservicesmayseemtobeunusualascontextforanegotiationanalysis, it is

highly relevant application (de Groot and Hermans, 2009). Writers concerned with providing

negotiationadviceoftenspeakoftheneedtonegotiateabouttheprocessaswellassubstantive

outcomesofES conservation schemes (Fisheretal., 1991).The structureof thenegotiation in

what follows is more along the lines of institutional design than a more standard textbook

negotiation problem, such as negotiating the price of an asset; nevertheless it is a new and

potentiallywidelyapplicablecontext.

84

4.4Methods

Negotiationanalysisrequiresconstructionofanegotiationtemplate(Raiffaetal.,2002),builton

an understanding of motivations, objectives, and characteristics of acceptable alternatives,

identifiedseparately foreachofthenegotiatingparties.Theoverallprocedureusedtodevelop

thenegotiationtemplateinthisstudyisshowninFigure1.

Figure 1: Research method: steps to build and evaluate the negotiation template for the design of mechanisms to improve ES provision.

Herewe adopt Keeney’s value‐focused thinking (VFT) as one important basis for creating this

template(KeeneyandRaiffa,1991;1993).VFTinvolvesaprocessofclarifyingvaluesimportantto

decision makers and interested parties. That information can then be used to identify

requirementsofadditionalinformationfordecisionsandforservingasabasisforcreatingmore

attractivealternatives(Keeney,1992;GregoryandKeeney,1996).VFThashadwideapplication

in a variety of contexts, including environmental management, business strategy, defence,

engineering,andplanning,particularlyforcitizens’involvementincomplexdecisions(McDaniels

et al., 1996) and recently in negotiations for PES (de Groot and Hermans, 2009). Keeney and

Raiffa (1991) have discussed value‐focused thinking for both parties in a negotiation as a

fundamental step indevelopingadecision template,which isan important step innegotiation

analysis.

Ourmethod is informedby twoclosely related approaches,namely: discourse‐basedandArea

DevelopmentNegotiations.Thefirstapproach involvestheuseofsmallgroupsofstakeholders’

representatives to deliberate on relevant aspects of ecosystem services, bringing to the

discussiontheirvalues,attitudesandbeliefs(McDanielsetal.,1999;WilsonandHowarth,2002).

85

The secondapproach includesa sequenceof steps tobuildunderstandingof thecase through

interviews,focusgroups,expertinformationandexperientialencounterswiththecasestudyasa

way to identify common strategies in the presence of divergent interests (Scholz and Tietje,

2002).

86

4.4.1DataAcquisition(Step1)

Steps1isconcernedwitheliciting,byinterviewsandfocusgroupswitheachstakeholder,value

perspectivesandotherrelatedcognitiveinformationrelevantforthedecisioncontext(detailsin

section4.1.1and4.1.2).

The discussions with actors were structured to elicit the participants’ values for actions and

decisionsregardingwatershedmanagementeffortstoreduceerosion.Toelicitvaluesweasked

stakeholderstocommentonwhatisimportanttothemregardingdirectpaymentstofarmersas

ameans to foster conservation in upstream farms, drawing on the analogy of an existing PES

scheme inCostaRica (Pagiola,2008). Specificquestionselicited their viewson implementinga

scheme for soil conservation in productive farms and the associated constraints. We asked

questionsonwhatmakesidealalternativesolutionsandtheirexpectedbenefitsinordertoelicit

theirviewsonalternativestrategies.Thefocusgroupdiscussionsweretranscribedandanalyzed

toidentifyandstructurekeyobjectivesinordertousethemasinputsintheconstructionofthe

negotiationcriteriaonaSCP.

4.4.1.1Datafromfarmers

Interviews. In June2007, individual semi‐structured interviewswereconductedwithproducers

( 50)n = from theBirriswatershed, selected randomly from the producers’ census of the local

agriculturalextensionoffice.Wecompletedacomprehensivereviewofavailablefarmpractices

for soil conservation and their adoption in the area based on existing documents in the

agriculturalextensionoffice.Inpart,thisreviewhelpedidentifykeybarriersandopportunitiesfor

the successful implementation of conservation practices in the Birriswatershed. The interview

instrument included questions to elicit farmers’ perceptions on the health of their soil and

whether they believe they face problems with soil degradation or erosion. We also included

questions to capture their understanding of erosion control processes using thementalmodel

method discussed in Morgan et al. (2002and open‐ended questions about the factors

constrainingorpromotingtheimplementationofsoilmanagementpractices(detailedresultscan

befoundinVignolaetal.,2009).

87

Focus group. To build on the individual interview results, we conducted a focus group with

representativesofproducers’associationsintheofficeoftheBirrisagriculturalextensionservice

in order to discuss the objectives for soil conservation programs. The participants included

representatives of a large local cooperative of producers, a member of a national producers’

association,amemberofasmallproducers’association,andanindependentproducerwhomisa

localleadproducerpromotingsoilconservationonproducers’farms.

Webeganwithanopendiscussiononthestateofsoilerosioninthewatershedandthefactors

thatmake itvulnerabletoerosion(i.e.climate, landusepracticesand infrastructure).Next,we

explored past experiences regarding conservation of SRS and the alternatives that have been

considered, in conjunctionwith incentives that would be helpful to foster their adoption.We

then employed methods discussed in Keeney (1992), and related papers, to elicit concerns,

constraints,andvalueperspectives.

4.4.1.2DatafromJASEC

Interviews.Aseriesofinformaldiscussionswasheldwithdifferentindividualsworkingwithinthe

JASEC‐Birrishydropowerproductionsystem.Those interviewed includedthevice‐president, the

manager of the hydropower facility, and the individuals responsible for watershed operations

andforsedimentmonitoringinthedam.Thebroadintentwastolearnaboutsystemoperations,

the effects of siltation on operations, and the utility’s interests and concerns in considering

possibleESprogramstoaddresssiltation.WeagainfollowedmethodsoutlinedinKeeney(1992)

tostructurethediscussion,asabasisforclarifyingthefundamentalendsandmeanstoachieve

theoverallgoalofaSCP.

FocusGroup.Tobuildontheresultsofthe interviews,afocusgroupwasheldwithfoursenior

officialsfromJASEC,including:thevice‐president;thecoordinatoroftheElectricOperationUnit;

the director of the Special Project Unit; and the coordinator of an environmental education

programintheBirriswatershed.Participantsdiscussedoncurrentsoilconservationactivitiesand

whatarethecharacteristicsofidealalternativestoimproveSRSprovision.Positiveandnegative

aspectsof currentuser‐financedPESprogramsandpotential alternativeshelpedcharacterizing

JASEC’smeansandendstoachieveadesirableSCP.

88

4.4.2Synthesisfortemplatebuilding(Step2)

Step 2 involves synthesizing and analyzing transcribed interviews and focus group discussions

which complementary information for evaluation of natural resources (Kaplowitz and Hoehn,

2001).HereweusedtheWITI(“Whyisthatimportant?”)methodofquestioningininterviews,as

discussedinKeeney(1992)andClemenandRielly(2001),todistinguishamongendsandmeans

relevant for the two groups of decision makers (farmers and the electric utilities). For a

negotiation template, the ends or fundamental objectives are used as the basis for criteria to

compare and evaluate alternatives from each party’s viewpoint, while themeans are used to

createbetter,moreattractivealternatives.Basedonthisinformationwepreparedadraftsetof

objectives,andanobjectivesnetwork,alsocalledameans‐endsnetwork(Keeney,1992).Step2

also involves creating attractive alternatives to be addressed in the negotiation. Gregory and

Keeney(1994)andKeeney(1992)articulatehowstakeholdervaluescanbeusedtohelpcreate

more attractive alternatives through a process of considering and building creative linkages

amongmeansandends.

Thisvalue‐orientedinformationonalternativeswascomplementedbyexpert‐basedsources.We

draw on literature regarding watershed management and soil conservation mechanisms to

identify various aspects or components of a mechanism for soil conservation, as well as the

alternativeoptionsavailable.Weuseanapproach referred toasastrategy table (Clemenand

Reilly, 2001) to graphically communicate the structure of the alternatives,which includes four

separatecomponents(discussedbelow)andseveralsub‐alternatives.Thesesub‐alternativescan

thenbecombinedinvariouswaystocreateseparatestrategiesofamechanismforfosteringES

provision,whicharethealternativesconsideredinthistemplate.Inotherwords,combiningsub‐

alternatives from each of the four components provides the basis for an array of different

strategies,whichcould serveas thebasis fornegotiationandultimatelyagreementamong the

twopartiestocreateandstructureasoilconservationprogram.

4.4.3Verificationandevaluationofalternativestrategies(Step3)

In Step 3, the two groups are asked, in two separate focus groups, to first verify and then

evaluate the alternatives under each of the components of mechanism design. We verified,

followingHabron et al. (2004), the diagramswedrewbased on stakeholders’ consultations in

89

ordertoconfirmwehadadequatelycapturedparticipants’perspectiveontheproblem.Hencein

eachofthesemeetings,our interpretationsoftheresults fromearliermeetingswerereviewed

andverifiedbyeachpartytothenegotiation.

Afterverificationbythetwogroups, theelicitedendscanserveastheobjectivesorevaluation

criteria for each group in the decision context12. The evaluation process is akin to eliciting the

‘zoneofbargaining’(Raiffaetal.,2002)foreachofthecomponents13.Thetwogroupswerealso

separately asked to identify which sub‐alternatives under each of the four components of

strategies are acceptable from their viewpoint. The intent of this process is to define the

negotiation space, or areas of potential agreement, that could help shape and direct the

subsequentnegotiationprocess.

We did this by asking the individuals in each group of participants to review the specific

components, and sub alternative under each component, as presented in aworkbookwehad

previouslyprepared.Participantswereasked to indicatewhichalternativeswereacceptable to

them(atasksimilartoapprovalvoting;McDanielsandThomas,1999).Wealsoaskedparticipants

to indicatewhich alternativewould bemost preferred in negotiation.We tried to ensure the

questions were understood by all participants. The judgment tasks were kept as simple as

possible to facilitate participation14, but elicited important information regarding initial

perspectivesonaspectsoffuturenegotiations.

4.5Results

4.5.1Objectivesofnegotiatingparties

4.5.1.1Farmers

Themeetingwithfarmers ( 8)n = washeldattheBirrisagriculturalextensionofficeinFebruary,

2009,andincludedleadersoffarmers’associationsandindividualproducers.Figures2presents

12Objective statements requireadecisioncontext, anobject, andadirectionofpreference (Keeney,1992). Forexampleoneobjectiveisstatedasstatedasfosteringorimproving(directionofpreference)farmincomes(object),inthecontextofimplementation of a soil conservation program (context). Sub‐objectives (e.g. short term (less than two years) and longterm(twoyearsormore))furtherspecifytheobjectivestatement.13Notethatbecausethecomponentsarelargelyseparable,theycanbeconsideredandnegotiatedindividuallyaspartofalargeagreement.14TheBirrisagriculturalextensionofficerattendedthefinalworkshoptoanswerquestionsandprovidehelptofarmerswhohaddifficultyreadingtheworkbook.

90

theobjectivesof farmersregardingthenatureofadesirablemechanismforanESprogram,as

derivedfromthemethodsdiscussedinSection4andverified inthefinalmeetingwithfarmers.

Ontheleftsideofthediagramarevariousmeansraisedindiscussions,whiletherightsideshows

the ends or fundamental objectives important to farmers in terms ofmechanism design. The

objectives listed on the right side reveal the characteristics of the ideal mechanism from the

farmers’viewpoints,whichinclude:1)provideincentivesforsoilconservation,2)ensureequityin

distribution of incentives, 3) promote learning and adaptive management (i.e.. about soil

conservationpractices)and4)achievetheoverallobjectivesofaSCP(i.e.thatworkwellforboth

farmers and JASEC in addressing soil conservation). Theremaywell be tradeoffs among these

objectives, or no alternatives that satisfy all at once, so care is required in designing and

evaluatingthealternatives.

Fig. 2. Farmers’ network of means-ends objectives for a desirable soil conservation program.

91

Themeanson the left side of Figure2 showkeyaspects for creatingattractivealternatives to

achievetheoverallgoalsofsoilconservationinthewatershedfromfarmers’perspectives.Note

that many of the means for designing a desirable SCP relate to fundamental objective of

achievingprogramsuccess.Inotherwords,onecriterionfordesigningthemechanismiswhether

itislikelytoachievethefundamentalendsofprogramimplementation15.

4.5.1.2JASEC

ThemeetingwiththeExecutiveBoardofJASEC ( 5)n = washeldattheircorporateadministrative

officeonMarch,2009.Participantsincludedallthosewhotookparttothefocusgroupsessions

forstep1plustheJASECresponsibleforsedimentmanagementandmonitoring.Fig.3(parallelin

structuretoFig.2)addressestheobjectivesofJASECforthedesignofamechanismforfostering

SRS provision. Again, we drew on the results of interviews and focus group with JASEC to

distinguishendsfrommeansandthenconfirmedtheseresultsintheJASECfinalworkshop.

15 Similar linkagesbetween fundamentalobjectives for the process, and for theactual implementation,arefoundinKeeney,McDanielsandRidge‐Cooney,1996

92

Fig. 3. JASEC’s network of means-ends objectives for a desirable soil conservation program.

In Fig. 3, in the right column, we find four objectives that are related to the fundamental

achievements of a Soil Conservation Program that show concerns in different dimensions. For

example, JASEC is concerned with minimizing investment needs by targeting areas for SRS

provisionprovidinghigherreturnininvestments.Moreover,theyexpressedthewishtoimprove

knowledgeoncostsandbenefitsofsupportingaSCPwhilemaintainagoodpublicimagethrough

thepromotionofaSoilConservationProgram.AcomparisonbetweenFig.2and3showsthatthe

two sets of parties to the negotiation share some objectives, and also have some differing

concernsovertheaspectsofadesirableSCP.That isnotunexpected,and it ishelpful tomake

thesedifferencesexplicit(Sebenius,1992).

4.5.2Alternativesasstrategiesacrossseveraldimensions

93

Thealternatives(whicharestrategiescomposedofsub‐alternativesundervariouscomponents)

consideredforthenegotiationtemplatearebasedonthecomplementaryuseofthestructured

means‐ends networks in Fig. 2 and 3 and expert‐based literature. Table 1 shows the four

components important in designing a mechanism for a soil conservation program, the sub‐

alternativesundereachcomponent,andtheinformationsourcesusedtoidentifythem.Thefour

componentsofdesigningamechanismforESprovision(Engeletal.,2008;Wunderetal.,2008):

include(i)SelectionofproducerstoreceiveSCPsupport,(ii)theNatureofsupportthattheSCP

delivers to producers, (iii) the Nature of the contract establishing a formal linkage between

providersandbeneficiariesofSRS,and,finally,(iv)Whoshouldbeintermediarybetweenthetwo

partiesintheSCP.

94

Table 1. Components of alternatives drawn from participants and expert-based literature.

Component Alternatives SourceAlloverthewatershed Participants(farmers,utilities)

Bidthroughauctions Ferraro,2008.

Highpriorityareas Goldmanetal.,2007.

Selectionofproducers

Socioeconomicstatus Pagiolaetal.,2005.

Random Participants

Credit ParticipantsNatureofsupport

de‐taxation Participants

Technicalassistance Participants

DirectPayment Engeletal.,2008;Wunderetal.,2008;Pagiola,2008.

Flexible BougheraraandDucos,2006.Natureofcontract

Rigid BougheraraandDucos,2006.

None Participants

Whoshouldbeintermediary Eachsingleproducer Participants

Agriculturalextensionoffice Participants

Watershedcommittee Participants

Producer'sassociation Participants

The sub‐alternativesundereachcomponentwerederived fromboth stakeholders’ andexpert‐

based literature. For example, the sub‐alternatives under the component “Selection of

producers” are the result of both actors’ consultation and literature. Here, farmers indicated

concern over “Characterizing farmers by socioeconomic characteristics” as a prerequisite of

targetingincentivesinequitablemannerwhileJASECexpressedconcernover focusingselection

of farmers in “Highpriorityareas”.Both these sub‐alternativesare complementedby the sub‐

alternative to select farmers by an auction‐bidding process (Ferraro, 2008).On the other side,

95

based on interviews with farmers and JASEC we could identify that the current strategy for

selectingfarmersforsupportisat“Random”16.

Similarly, for thecomponent “Natureof support”, literaturewidely supports theuseof “Direct

payments”tosupportfarmers’effortsinsoilconservation(Engeletal.,2008).Ourinterviewsand

focus groups additionally identified the potential use of indirect economic incentives, such as

removing taxes on agricultural inputs (i.e. “De‐taxation”), “Credits”17 for soil conservation, and

the provision of ”Technical assistance”18. For the component “Nature of the contract” to

formalizeagreementbetweenthenegotiatingparties,wefoundastatus‐quoalternative(i.e.”No

contract”) from stakeholder consultations. Additionally, two other sub‐alternatives from the

literature (Bougherara and Ducos, 2006), namely “Flexible” (defined annually) or ”Rigid”

contracts (fixed over a period of several years) were included to identify potential trade‐offs

between the concerns of JASEC and farmers. More specifically, farmers attempt to maintain

flexibility, to allow ongoing adaptation to changingmarkets and climate conditions; therefore

theyare likelymoreattracted by thepossibility of flexible contracts. For JASEC, rigid contracts

mightbepreferable,toensurethatpractices implementedatthestartofacontractperiodare

maintainedthroughoutthedurationofthecontract(i.e.eveninthefaceofchangingmarketor

climateconditions).

Finally, the alternatives for ”Who should be intermediary” reflect concerns over which

institutionalentity isorshouldbeentitledtosupportactivitiessuchascontracting,monitoring

and following‐up of soil conservation practices.While entities such as ”Agricultural extension

office”, “Producers’ associations” and “Individual producers” are known to both groups,

“Watershedcommittee”isfamiliaronlytoJASECwhosewatershedmanagementplanincludesa

mandate to create such a committee. Sub‐alternatives under the component “Selection of

farmers” and “Nature of support” are not mutually exclusive, and could be combined. For

example, for “Selection of producers”, targeting farmers in priority areas and letting them

compete throughauctions couldbothbe selected. Similarly, for ”Natureof support”,providing

16Although,technicalassistanceisfocusingonbroadly‐definedareaswithseriouserosion,thereisnoestablishedprocessfordefiningpriorityareas.17Creditswerementionedinreferencetopastexperienceintheareawherebankswereprovidinglow‐rateloanstoselectedproducersusingaspecificvarietyoflow‐impactpotatocrop.18Technicalassistancewasdefinedtoincludeon‐goingsoilconservationprograms.Extensionofficersvisitfarmsperiodicallytoadviseonandmonitorfarmers’implementationofsoilconservationpractices.Alsotrainingisprovidedifnewtechniquesareintroduced.

96

“Directpayment”couldbecombinedwith“Technicalassistance”.Ontheotherhand,alternatives

under “Nature of contracts” and “Who should be intermediary”, sub‐alternatives aremutually

exclusive.

97

4.5.3Acceptability,preferredalternativesandzonesofbargaining

Intheevaluationphase,participantsinthefocusgroupusedaworkbookorganizedtoevaluate

the components one at a time. Fig. 4 summarizes the consultation results of the evaluation

processinasimilarwayaspresentedinScholzandcolleagues(2007).Inseparatemeetingswith

bothfarmersandJASEC,weelicited,whichofthealternativesundereachcomponentwouldbe

acceptable(i.e.morethanonesub‐alternativecouldbeselected),andwhichwasmostpreferred.

Thefigureshowsalternativesacceptabletoatleastsomeofthefarmers,tosomerepresentatives

of JASEC, and to both. To avoid an arbitrary aggregation rule, we report the percentage of

participantsinagreement,ineachboxofthetable.

Fig. 4. Acceptable and most preferred alternatives for farmers and JASEC (note: numbers in the bottom-right (left) of boxes represent the percentage of votes given by JASEC (farmers) to that sub-alternative).

Resultsfromthenegotiationexerciseshowthatthecharacteristicsoftheprogram,asdefinedby

stakeholders’ preferences, can serve the learning requirements for both parties and can be

efficient in the provision of SRS to both hydropower and farmers. More specifically, such a

programcould,consistentlywiththeirpreferences,concentratein“Highpriorityarea”,promote

“Technical assistance” instead of PES, use either flexible or rigid contracts and entitle

98

“Agricultural extension office” or the not‐yet‐created “Watershed committee” as intermediary

betweentheprogramandthefarmers.Somesurprisesarisefromtheseresults.Indeed,farmers

showlowacceptanceforhavingtheprogramspreadalloverthewatershed(i.e.theysuggestto

concentrate in “High priority areas”) although preference for program to cover the whole

watershedmightbeexpected.Similarly,monetaryincentivessuchas”Credit”and“De‐taxation”

or ”Direct payment” (accepted partially only by JASEC), have, contrasting expectations, low

acceptanceamongfarmers.Ontheotherside,farmers,consistentlywiththeirvalues,expressed

lowacceptance for selectionmechanisms thatpromotecompetitionamong themsuchas “Bid

throughauctions”.Forthecomponent“Natureofthecontract”,bothparties’acceptanceofrigid

and flexible contracts as alternatives to status quo (i.e. no contract) underlies a shared

perception of the need to establish a formal agreement to reduce soil degradation in the

watershed.While for JASEC no contractmight represent uncertain investment, for farmers no

contract might represent uncertain access to support. It is interesting to note that sub‐

alternatives supporting their direct or indirect (i.e. through their associations) involvement in

negotiating supportwith the programreceived little support contrary toexpected.Both JASEC

andfarmersacknowledgedtheroleofagriculturalextensionofficeorwatershedcommitteefor

intermediation.

Finally,aftertheevaluationexercise,bothfarmersandJASECwereaskedtoexpresstheirviews

onpossiblenextstepsinordertomoveforwardthisnegotiationprocess.Bothgroupsarticulated

theneedtoidentifyhighriskareas,andbetterspecifythenatureoftechnicalassistancesought

by farmers.Moreover, both also indicated the need to create a structure for developing and

implementingcontracts,perhapsthroughanewwatershedmanagementprocess.Farmersalso

stressedtheneedtoengageintheprocessotherfarmerswithinhigh‐riskareas.

99

4.6Discussion

Our paper builds on the recent paper of de Groot and Hermans (2009) whom used a

stakeholders’value‐basedanalysis tooutlinepotentials forPESnegotiationsbetweenproviders

anduserswater‐related ES.Our focus, however differ for three reasons. First,we focusedour

attentiononthebuildingofaninstitutionalmechanismratherthanmainlyontheeffectiveness

andstructureofdirectpaymentstofarmers.This isespeciallyrelevantwhenconsidered inthe

CostaRicancontextwherei)pioneeringexperienceonimplementationofPESschemeshasput

mostemphasisonthiscompensationmechanismthanonthevaluesand interestsofproviders

andusersofES,andii)thelocationinCentralAmericacanprovideinsightsonnegotiationsforES

contextthatmightberelevantfordevelopingcountriescontext.Second,weputemphasisison

the provision of water‐soil regulation services that provide potential for joint gains between

farmersandhydropowerratherthanputtingemphasisontheprovisionofservicesthatbenefit

principally itsuser (e.g. thewaterboard).Third,differently fromdeGrootandHermans(2009)

weusedvalue‐basedandexpert‐based literature tobuilda negotiation template, validateand

evaluate it to test ourmethodwith stakeholders’ preferences. Results from our consultations

withstakeholdersshowthatthestructureofmeans‐objectivesexpressesconcernsforeconomic

(i.e.transactioncostsofmonitoringimplementationofpractices,definecontractsanddistribute

incentives)andproceduralaspects(i.e.meetlegalrequirementsinsigningcontractsandprovide

incentives)ofthedesirablemechanismforsoilconservation.Thefiguresandthefinaltemplate

evaluationinthispaperprovideabasisforstructuringthefuturenegotiationprocess.

Thecomponentsandsub‐alternativesreflect,beyondavailabilityandconsistency19ofresources

for the program, concerns (McKenzie, 1997; Hajkowicz, 2009) on procedural aspects, such as

definingstrategiesfordistributionofresourcesandmonitoringofprogramimplementation,are

important.Drawingalternativestrategiesfromresultsofbothopendiscussionswithparticipants

(i.e.means‐endobjectivenetwork)andexpert‐basedliteratureallowswidening“thelimitedand

fallible perspective of a single actor by drawing on other experience and knowledge” (Smith,

19 As outlined in Hajkowicz (2009), many watershed conservation program have not yet adequately addressed whether

investmentsinresourcesareconsistentwithfull‐rangeofcostsandbenefitsofSCPprovidingSRS.giventhefocuson“hard‐to‐quantifyandhard‐to‐monetize”environmentaloutcomes.

100

2001). For example, for the component “Nature of the contract” between providers and

beneficiaries of the ES stakeholders had not mentioned important aspects of the possible

contracttypebetweentheprogramandfarmers.Bringingnewknowledgeinthenegotiationof

themechanism,expertliterature(BougheraraandDucos,2006)showsthattheextenttowhich

the“Flexibility”ofthecontractallowsfarmerstoadaptsoilconservationpracticesovertimecan

beanacceptedalternativetoovercomestatus‐quo,whensubjecttostakeholders’evaluation.

Ourstrategytable,basedoninputsfrombothconsultationswithparticipantsandtheliterature,

provides a set of alternatives under each component, some of which are acceptable or most

preferredtobothgroups20.Forexample,bothfarmersandJASECpreferthattheprogramselect

farmersbasedontheprincipleoftargetinghighriskareas(steepareaswithexposedslopes).This

convergence of perspectives on targeting priority areas for the ES provides an opportunity for

increasingefficiency in hydrologicalES schemes (Goldmanetal., 2007;ParkhurstandShogren,

2007).Moreover,targetingofESpriorityforwatershedconservationwouldimproveCostaRica’s

currentPESprogram,(Wunscheretal.,2007;Pagiola,2008).

Someproceduralaspectssuchas“Selectionofproducers”haveprovidedcontroversialoutcomes.

Forexample,thealternative“Auctions”thathavebeenproposedasastrategyforconservation

programs to make farmers reveal their true opportunity costs in the provision of ecosystem

services(Ferraro,2008)havereceivedlittlesupport.AsFerraro(2008)notes,selectionbasedon

this self‐revealed opportunity cost can be seen as unfair by farmers, due to incentives for

competition among farmers with different resource endowment. Additionally, this sub‐

alternative requires the SCP to dispose of additional resources to cover the transaction costs

needed to runandmonitor thebiddingprocess. For thecomponent “Natureof support”,both

groupsexpressedpreference for “Technicalassistance” rather than“Directpayments” thatare

beingpromotedbynationalPESprograms.Thispreferencehighlightstheimportanceoflearning

over time,which ishighly relevant in theuncertain context inwhich thepartiesareoperating.

Mobilizing effective scientific knowledge and the appropriate technology (through technical

assistance)canbeaveryeffectivestrategyforsustainablelandmanagement(Cashetal.,2003).

This is especially the case in the context of the Birris watershed where neither the best

20 Some 240 strategies are possible (5x4x3x4), ignoring the potential for combining alternatives under the nature ofincentives.

101

conservation practices to reduce soil erosion nor the potential benefits for agricultural

productivity are precisely known by the parties. Moreover, this is particularly relevant

consideringthatuncertainclimatedynamicsandincreasesinextremeprecipitationeventsinthe

future might require continuous adjustments (i.e. learning over time). Farmers’ preferences

suggestedtheirconcernsthat“Directpayment”would involverestrictions(i.e. legal liability)on

soil management. In light of the small farm‐area available and the dependence on unstable

market prices, farmers consistently showedpreference for less‐binding solutions. As shownby

ZbindenandLee(2005), themajorityofparticipants inthecurrentlyoperatingCostaRicanPES

arelargelandownerswithrelativelyloweropportunitycostsforadoptingthelandusesrequired

byPEScontracts.Thus,unlessBirrisfarmersalsoperceiveprivatebenefitsfromsoilconservation,

thefixedpaymentmightbeperceivedastoosmall(Pagiola,2008)orunabletomatchunstable

agricultural market prices during contract period. JASEC, on the other hand, viewed direct

paymentsasacceptable,consistentwiththeirknowledgeonPESexperiences(bothgovernment

anduser‐financedPESschemes)inthecountry(Engeletal.,2008;Pagiola,2008).

Therelativelyhighdegreeof farmers’acceptanceof“Rigid”contractsdeservesmoreattention,

especially in the light of the agricultural production context. Based on their actual experience,

farmersmayhaveevaluatedthe“Natureofcontract”asawaytoensureacontract‐basedaccess

to“Technicalassistance”butwithoutthelegalimplicationsofreceivingdirectpayment.Technical

assistance is currently provided without a legal contract (i.e. no formal commitment) but the

provisionofthisserviceisdiscontinuousandwithalimitedgeographicalcoverage.

Whenaskedtoevaluatealternativesonthecomponent“Whoshouldbeintermediary”between

theprogramandfarmers (e.g. includingactivitiessuchasmonitoringandfollow‐upofprogram

activities),theformerexperienceofparticipants21mightplayanimportantrole(Imperial,1999).

Forexample,neitherofthegroupshaveexperiencewiththealternative“Watershedcommittee”

(representing relevantstakeholderssuchas farmers)because,thecommitteehasnotyetbeen

formed,andonly JASEChasacknowledged its formation in theirBirriswatershedmanagement

plan. Current positive experience with the agricultural extension office might underlie the

21Neitherofthegroupshaveexperiencewiththealternative“Watershedcommittee” (representing relevantstakeholders

such as farmers) because, the committee has not yet been formed, although it is included in the existing watershedmanagementplan.

102

convergenceofpreferencesoffarmersandJASECparticipants.Thelackofsupportforproducers’

associationsbybothgroupsmightpresumablybedue to concernsoverprincipal‐agency issues

andthelackoftrustintheseorganizations,factorswhichhavebeenalsoobservedbyahistorical

analysis of the farmers’ cooperative movement in the country (Anderson, 1991). The

“Agriculturalextensionoffice”,highlyappreciatedforthetechnicalassistanceservicesitprovides,

mightbeperceivedasatrustworthyandconvenientchannelforadministrativetasksneededto

jointheprogram.JASEC’slackofsupportfor“Eachsingleproducer”asintermediatorsmightbe

relatedtotheconcernsovertransactioncosts,indicatedbyexperiencesinotherregions(Engelet

al., 2008; Wunder et al., 2008). In sum, intermediary organizations, rather than direct

intermediation, linking the providers and consumers of ES is seen as a desirable aspect of the

program,asshowninananalysisofVoluntaryAgreementsinwatershedprotectioninCostaRica

(Mirandaetal,2007).Aninterestingpointraisedbyfarmerswastheperceivedneedtoinvolve

most farmers in targeted areas. Indeed, the efficacy of farmers’ soil conservation practices in

providing SRS depends in part on topographical gradients and on how their actions are

coordinatedwithupstreamanddownstreamneighbouringproducers.Thisterritorialconcernhas

receivedattentionalsoinrecentliterature(Goldmanetal.,2007;ParkhurstandShogren,2007)

thatdiscussedthepotentialsofappropriatetargetinginESprogramsusingcooperativebonuses

anddeclarationofenvironmentaldistrictsasameanstopromotefarmerscooperationinpriority

areas.

Finally,theidentifiedareasofagreementcanbeusedtocreatemanyalternativestrategiesthat

could be evaluated on the basis of stakeholders’means‐ends objectives. For example, if after

further negotiation, farmers would ultimately find direct payments acceptable in place of

reduced input taxes or credit, then some eight different strategies could be created from the

alternativesofmutualagreementfromTable4(1x2x2x2).Theseeightstrategieswouldneedto

be evaluated and negotiated by both farmers and JASEC to select the best overallmechanism

acceptabletoboth.

4.7Conclusions

Thisarticle focusedon thebuildingofalternative strategies foraESmechanismusingdecision

and negotiation analysis to explore stakeholders’ values and interests. We elicited the

103

subjectivelyperceivedcausalitiesofsoilerosionbyapplyingmentalmodelinterviews,thevalues

attachedtoitscontrolandthoseofrelevanceforthedesignofcollectiveresponsestrategiesby

applyingavalue‐focusedapproach.Stakeholdersanalyzedalternativesforbuildingamechanism,

ratherthanrelyingonpre‐establishedandfixedalternativeoutcomes.Thisapproachpermitted

ustoidentifyfundamentalvaluesofrelevantstakeholders(i.e.providersandbeneficiariesofES)

and zonesofagreement thatmayhelp shape thedesignofmore sociallyandenvironmentally

efficientfutureprograms.

WhilethemajorityofliteratureonPESassumesthatdirectmonetaryincentivesareappropriate

for fostering ES provision, this paper shows that in analyzing the complex institutional

requirementsof soil conservationprograms,other considerationsmatter. Indeed, stakeholders

base their preferences on issues such as fairness, trust in and experience of interaction with

others,andaccesslearningovertimeonthechangingcausesandsolutionsofthedegradationof

ES.Acknowledging these fundamentalaspects in thedesigningofamechanismmight increase

thewillingnessofinterestedpartiestofindagreementonSCP.Thisinturncouldimprovesocial

efficiencyandenvironmentalefficacyofconservationprograms.Furtherresearchonnegotiations

should include willingness among farmers to find agreements on the best mechanisms for

cooperation, especially in high priority areas where returns on investments in conservation

effortspromisetobegreater.

4.8References

Aguilar,E.T.,PetersonC.,Ramırez,P.,Frutos,R.,Retana,J.A.,Solera,M.,Soley,J.,GonzalezGarcıa, I.,Araujo,R.M., Rosa Santos, A., Valle, V.E., Brunet, M., Aguilar, L., Alvarez, L., Bautista, M., Castañonp, C.,Herrera,L.,Ruano,E.,Sinay,J.J.,Sanchez,E.,HernandezOviedo,G.I.,Obed,F.,Salgado,J.E.,Vazquez,J.L., Baca,M.,Gutierrez,M.,Centella, C.,Espinosa, J.,Martınez,D.,Olmedo, B.,OjedaEspinoza,C.E.,Nuñez, R., Haylock, M., Benavides, H., Mayorga,R., 2005. Changes in precipitation and temperatureextremesinCentralAmericaandnorthernSouthAmerica,1961‐2003.JournalofGeophysicalResearch,Vol.110,D23107.

Anderson, L., 1991.Mixedblessings: disruption and organization amongpeasantUnions in Costa Rica. LatinAmericanResearchReview26(1),111‐143.

Bougherara,D.,Ducos,G.,2006.Farmers’PreferencesoverConservationContractFlexibilityandDuration,AnEstimation of the Effects of Transactions Costs using Choice Experiment, , Working Paper, INRA,Rennes,France,26pp.

Cash, D., Clark, W.C., Alcock, F., Dickson, N.M., Eckley, N., Guston, D.H., Jäger, J., Mitchell, R.B., 2003.Knowledge Systems for sustainable development. Proceedings of the National Academy of Science(PNAS)100(14),8086‐8091.

Clemen,R.T.,Reilly,T.,2001.MakingHardDecisions,Duxbury,PacificGrove,CA.

104

deGroot,R.B.A.,Hermans,L.M.,2009.Broadeningthepicture:negotiatingpaymentschemesforwater‐relatedenvironmentalservicesintheNetherlands.EcologicalEconomics68,2760‐2767.

Dietz,T.,Ostrom,E.,Stern,P.C.,2003.Thestruggletogovernthecommons.Science302(5652),1907‐1912.Engel,S.,Pagiola,S.,Wunder,S.2008.Designingpaymentsforenvironmentalservices intheoryandpractice:

Anoverviewoftheissues.EcologicalEconomics65(4),663‐674.Ferraro, P., 2008. Asymmetric information and contract design for Payment for Environmental Services.

EcologicalEconomics65(4),810‐821.Fisher, B., Turner, R.K., Morlina, P., 2009. Defining and classifying ecosystem services for decision making.

EcologicalEconomics68,643‐653.Fisher, R.,Ury,W.,Patton, B.,1991.Getting to yes: negotiatingagreementwithout giving in. Eds.Houghton

MifflinHarcourt,200pp.Goldman,R.L.,Thompson,B.H.,Daily,G.C.,2007.Institutionalincentivesformanagingthelandscape:including

cooperationfortheproductionofecosystemservices.EcologicalEconomics65,333‐343.Gregory,R.,Keeney,R.L.,1994.Creatingpolicyalternativesusingstakeholdervalues.ManagementScience40

(8),1035‐1048.Guo, Z., Xiao, X., Li, D., 2000. An assessment of ecosystem services: water flow and hydroelectric power

production.Ecologicalapplications10,925‐936.Habron,G.B.,Kaplowitz,M.D.,Levine,R.L.,2004.AsoftSystemsApproachtowatershedmanagement:aRoad

Saltcasestudy.EnvironmentalManagement33(6),776‐787.Hajkowicz,S.,2009. The evolutionofAustralia’snatural resourcemanagementprograms:Towards improved

targetingandevaluationofinvestments.LandUsePolicy26,471‐478.Hammond, J., Keeney, R.L., Raiffa, H., 1999. Smart Choices: A Practical Guide to Making Better Decisions,

HarvardBusinessSchoolPress,Cambridge,MA.Harsanyji, J.,1982.SubjectiveProbabilityandthcTheoryofGames:CommentsonKadaneandLarkey'sPaper.

ManagementScience28(2),120‐124.Imperial,M.T.,1999. InstitutionalAnalysisand Ecosystem‐BasedManagement: The InstitutionalAnalysisand

DevelopmentFramework.EnvironmentalManagement,24(4),449‐465.Innes, J.E., Booher, D.E., 1999. Consensus building and complex adaptive systems. Journal of The American

PlanningAssociation65(4),412‐423.Ison,R.,Röling,N.,Watson,D.,2007.Challengestoscienceandsocietyinthesustainablemanagementanduse

ofwater:investigatingtheroleofsociallearning.EnvironmentalScienceandPolicy10,499‐511.Kadane, J.B., Larkey,P.D.,1982.SubjectiveProbabilityandtheTheoryofGames.ManagementScience28(2),

113‐120.Kadane, J.B., Larkey, P.D., 1983. The Confusion of Is and Ought in Game Theoretic Contexts. Management

Science29(11),1365‐1379.Kahan, J.P., 1983. On Choosing Between Rev. Bayes and Prof. von Neumann.Management Science 29 (11),

1334‐1336.Kaplowitz, M.D., Hoehn, J.P., 2001. Do focus groups and personal interviews cast the same light on natural

resourceevaluation?.EcologicalEconomics36(2),237‐247.Keeney,R.L.,Raiffa,H.,1991.StructuringandAnalyzingValuesformulti‐partynegotiations,inYoungH.P.,(ed.)

1991.NegotiationAnalysis.UniversityofMichiganPress.AnnArbor.Keeney, R.L.,1992.Value‐focused thinking:apath tocreativedecision‐making.HarvardUniversityPress,432

pp.Keeney,R.L.,McDaniels,T.L.,1992.Value‐focusedThinkingaboutstrategicdecisionsatBChydro.Interfaces22,

94‐109.Keeney,R.L.,Raiffa,H.,1993,DecisionswithMultipleObjectives:PreferencesandValueTradeoffs,Cambridge

UniversityPress,Cambridge,UK.Keeney,R.L.,1994.Usingvaluesinoperationsresearch.OperationResearch42(5),793‐813.

105

Keeney, R.L., 1996. Value‐focused thinking: identifying decision opportunities and creative alternatives.EuropeanJournalofOperationalResearch92(3),537‐549.

Keeney, R.L., McDaniels, T.L., Ridge‐Cooney, V.L., 1996. Using values in planning wastewater facilities formetropolitanSeattle.JournaloftheAmericanWaterResources32(2),293‐303.

Kosoy,N.,Martinez‐Tuna,M.,Muradian, R.,Martinez‐Allier, J.,2007. Payments forenvironmental services inwatersheds:insightsfromacomparativestudyofthreecasesinCentralAmerica.EcologicalEconomics61(2‐3),446‐455.

Magrin,G.,GayGarcía,C.,CruzChoque,D.,Giménez,J.C.,Moreno,A.R.,Nagy,G.J.,Nobre,C.,Villamizar,A.,2007. Latin America. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution ofWorkingGroupIItotheFourthAssessmentReportoftheIntergovernmentalPanelonClimateChange,Parry,M.L.,Canziani,O.F.,Palutikof,J.P.,vanderLinden,P.J.,Hanson,C.E.,Eds.,CambridgeUniversityPress,Cambridge,UK,581‐615.

Marchamalo,M.,Romero,C.,2007.Participatorydecision‐makinginlanduseplanning:anapplicationinCostaRica.EcologicalEconomics63(4),740‐748.

McDaniels,T.L.,Gregory,R.S.,Fields,D.,1999.Democratizingriskmanagement:successfulpublicinvolvementinlocalwatermanagementdecisions.RiskAnalysis19(3),497‐510.

McDaniels,T.L.,Thomas,K.,1999.Elicitingpreferencesforlandusealternatives:astructuredvaluereferendumwithapprovalvoting.JournalofPolicyAnalysisandManagement18(2),264‐280.

McKenzie,S.H.,1997.TowardIntegratedResourceManagement:LessonsabouttheEcosystemApproachfromtheLaurentianGreatLakes,EnvironmentalManagement21(2),173‐183.

Millennium Ecosystem Assessment (MEA), 2005. Ecosystems and human well‐being: hydrological ecosystemservicesreport.IslandPress,WashingtonD.C.,USA.

Miranda,M.,Dieperink, C.,Glasbergen, P.,2007.Voluntaryagreements inwatershedprotection experiencesfromCostaRica.Environment,DevelopmentandSustainability9,1‐19.

Morgan,M.G., Fischhoff, B., Bostrom,A., Atman, C.J., 2002. Risk communication: amentalmodel approach.CambridgeUniversityPress,354pp.

Pagiola,S.,2008.PaymentsforenvironmentalservicesinCostaRica.EcologicalEconomics65(4),712‐724.Pagiola, S., Arcenas, A., Platais, G., 2005. Can Payment for Environmental Services help reduce poverty? An

explorationoftheissuesandtheevidencesfromLatinAmerica.WorldDevelopment33(2),237‐253.Parkhurst,G.M.,Shogren,J.F.,2007.Spatial incentivestocoordinatecontiguoushabitat.EcologicalEconomics

64,344‐355.Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., McNair, M., Crist, S., Shpritz, L., Fitton, L.,


Ponce, L.O., 2006. Economic instruments inwatermanagement: the caseof Costa Rica.Natural and EnergyResources,UnitedNationsEconomicCommissionforLatinAmerica(ECLAC),Mexico,p.59.

Raiffa,H.,1982.TheArtandScienceofNegotiation.BelnapPress,Boston.Raiffa,H.,Richardson,J.,Metcalfe,D.,2002.Negotiationanalysis:thescienceandartofcollaborativedecision‐

making.HarvardUniversityPress,548pp.Roth,A.,Schoumaker,F.,1983.ExpectationsandReputations inBargaining:AnExperimentalStudy,American

EconomicReview83(73),362‐372.Rothkopf,M.H.,1983.ModelingSemirationalCompetitiveBehaviour,ManagementScience29(11),1341‐1346.Sanchez‐Azofeifa, G.A., Pfaff, A., Robalino, J.A., 2007. Costa Rica’s Payment for Environmental Services

program:intention,implementationandimpact.ConservationBiology21(5),1165‐1173.Scholz, R.W., Tietje, O., 2002. Embedded Case Study Methods: Integrating Quantitative and Qualitative

Knowledge.ThousandOaks‐SagePublications,392pp.Scholz,R.W.,Stauffacher,M.,Bösch,S.,Krütli,P.&Wiek,A.(Eds.).(2007).EntscheidungsprozesseWellenberg

‐ Lagerung radioaktiver Abfälle in der Schweiz. ETH‐UNS Fallstudie 2006 [Decision processes

106

Wellenberg ‐ Radioactive Waste in Switzerland. ETH‐NSSI Case Study 2006]. Zürich, Chur: VerlagRüegger.

Sebenius,J.K.,1992.Negotiationanalysis:acharacterizationandreview.ManagementScience38(1),18‐38.Smith,G.,2001.Takingdeliberationseriously:institutionaldesignandgreenpolitics.EnvironmentalPolitics,10

(3),72‐93.Southgate, D., Macke, R., 1989. The downstream benefits of soil conservation in ThirdWorld Hydroelectric

Watersheds.LandEconomics65(1),38‐48.Vignola,R.,Koellner,T.,Scholz,R.W.,McDaniels,T.L.,2009.Decisionmakingbyfarmersregardingecosystem

services:factorsaffectingsoilconservationeffortsinCostaRica.SubmittedtoLandUsePolicy.Wilson,M.A.,Howarth,R.B.,2002.Discourse‐basedvaluationofecosystemservices:establishingfairoutcomes

throughgroupdeliberation.EcologicalEconomics41(3),431–443.Wossink, A., Swinton, S.M., 2007. Jointness in production and farmers’ willingness to supply non‐marketed

ecosystemservices.EcologicalEconomics64(2),297‐304.Wünscher,T.,Engel,S.,Wunder,S.,2008.Spatialtargetingofpaymentsforenvironmentalservices:Atoolfor

boostingconservationbenefits.EcologicalEconomics65(4),822‐833.Wunder, S., Engel, S., Pagiola, S., 2008. Taking stock: a comparative analysis of payments for environmental

servicesprogramsindevelopedanddevelopingcountries.EcologicalEconomics65(4),834‐852.Young,H.P.,1991.Negotiationanalysis.UniversityofMichiganPress,204pp.Zbinden,S.,Lee,D.R.,2005.Payingforenvironmentalservices:ananalysisofparticipationinCostaRica’sPSA

program.WorldDevelopment33(2),255‐272.

108

5. Governance across-scales: regulatory mismatches and the role of boundary organizations for soil regulation-services under climate change in Costa Rica

Authors:RaffaeleVignola,TimL.McDanielsandRolandW.Scholz.

Paperstatus:submittedtoJournalofGlobalEnvironmentalChange.

Abstract

The degradation of ecosystem services such as soil erosion control results from complex

interaction of climate extremes and soil management decisions. This complexity poses new

challenges to governance systems for the definition of strategies to adapt to these changing

conditions.Regulatoryframeworkandaccesstoinformationdeterminethecapacityofactorsto

achievetheirgoalsandcontributetosoilerosioncontrol.Inthispaper,weusethecasestudyof

Birris watershed where the degradation of Soil Regulation Services of upstream ecosystems

affects both upstream farmers and downstream hydropower dams. The Birris watershed, as

otherpartsofCentralAmerica,ischaracterizedbyhighlanduseconflict(duetohighslopesand

soil‐intensive agriculture) and is highly vulnerable to increasing precipitation extremes. We

analyzeformal regulatoryframeworkand information‐exchangenetworksthatcharacterizethe

constellationofactorsacrossscalesthatdirectlyorindirectlyinfluencetheprovisionanduseof

these ecosystem services. Our results show that two important regulatorymismatches across

scales affect the definition of societal responses at the local level (i.e.where direct actions to

promote adequate provision and use of ES happen). Additionally, through analysis of

information‐exchange networks among actorswe identify an important boundary organization

such as the watershed agricultural extension office that helps diffusing information from

differentscalesandpolicyareas.Theformalizationoftheboundaryroleofthisorganizationcan

revealastrategicstepinpromotingadaptivegovernance.Indeed,thiscanimprovethecapacity

of this organization to systematically recollect evidences of impacts on and responses to soil

degradation in the area. Adaptive governance can benefit from the dissemination of this

information to actors that play an important role in formulating regulations, allow and design

economic incentives and implement scientific research on the degradation of soil regulation

services.

109

Keywords:Ecosystemservices,adaptationtoclimatechange,adaptivegovernance, information

exchangenetworks.

5.1Introduction

TheMillenniumEcosystemAssessment(MEA)paintsagrimpictureof rapidgrowth indemand

for ecosystem services22,while two thirds of ecosystems providing services (ES) are in decline

(MEA,2005).Thepictureisevenmoredisturbingwhenthedisruptiveeffectsofclimatechange

areconsidered(Fischlinetal.,2007).Coordinatedadaptationefforts,withincomplexinstitutional

and governance contexts and under extreme uncertainties are required if ES supplies are to

bettermeetESdemandsunderthepressuresofclimatechange(Hulme,2005;BodinandCrona,

2009). Yet often i) those relying on a specific type of ES are spatially and administratively

differentfromii)thosewhoinfluencethequalityandquantityofthatESprovisionandiii)those

whogenerate,elaborateordisseminateinformationthatisrelevantformanagingthoseES.The

managementofSoilRegulationServices(SRS)providesacaseinpoint.Farmersonupstreamsites

clearlyhaveindividualincentivestoavoiderosionandretainsoilsontheirfarms,butoften lack

technical knowledge,equipment, capital, timeor cooperation of others fo needed inorder to

changepracticestoreduceerosion.Downstreamneighboursandmoredistantwaterusers,such

ashydroelectricelectric facilities,bear costs that couldbeavoided if formal regulationswould

allowmoreresources(e.g.financialandinformation)tobedevotedforupstreamerosioncontrol.

Intheparlanceofeconomists, thepotentialpositiveexternalitiesofSRS inavoidingsiltationof

reservoirsarenotinternalizedintodecisionsbyupstreamfarmers.

Theeffectsofclimatechangeareexpectedtoincreasetheintensityofrainfall,andthusincrease

erosivity,withevenmoresoil loss.Inthiscontext,actorsinvolvedinSRSmanagementdecisions

need appropriate information to guide their technical, administrative and socially‐sensitive

actions on ecosystems. However,most governance systems for ecosystems and adaptation to

climate change, however, are in their initial stages of recompiling information on impacts and

analyzingfeasibleresponses inthefaceofuncertainties (Eastaughetal.,2009). Inthiscontext,

22These include:provisioningservicessuchasfood,water,timber,andfiber;regulatingservicesthataffectclimate,floods,

disease, wastes, and water quality; cultural services that provide recreational, aesthetic, and spiritual benefits; andsupportingservicessuchassoilformation,photosynthesis,andnutrientcycling(MEA,2005,p.9)

110

research is needed to improve understanding of the role of key organizations in information‐

sharingnetworksasastrategicallyfordesigningadaptationresponses.

Information‐sharing among organizations is shaped by an existing set of formal and informal

institutionalstructuresandgovernancearrangementsthatdeterminethecapacityforcollective

responsestoESdegradationproblems(Ostrom,2007).Theseincludeformalgovernanceassetby

lawsandmandates,butalsoinformalinteractionsamongactorsbeyondtheirspecificmandates

andgeographicallocation.Thisgovernancestructureframes,overseesandimplementsresource

management policies from broader levels of governance, such as regional or national or

international,toindividuallandusersatthelocallevel(Adgeretal.,2005).

Recentstudiesofgovernanceforglobalenvironmentalchangehaveidentifiedtwocrucialaspects

foranalysisofsocietalcapacitytodesignandimplementadaptationstrategies.Thefirstconcerns

the interaction of actors at multiple scales of governance needing different detail‐levels of

informationonglobalenvironmentalimpacts(CashandMoser,2000).Often,mismatchesorgaps

ariseinregulatoryeffortsandgovernanceacrossscales,whichcanthwarttheireffectiveness.The

secondreferstotheroleofboundaryorganizationstoserveasconduitsthatprovidelinkages(i.e.

information and knowledge exchange), across scales and organizations belonging to different

communitiessuchasscience,government,andcivilsocietythatmakesdecisions influencingES

management(Agrawalaetal.,2001;Cashetal.,2003;Turtonetal.,2007).Todate,littleresearch

hasconsideredbothofthesethemesinonepaperusingempiricaldata,toseei)howquantitative

network analysis can identify key boundary organizations that may help build linkages across

scales, and how ii) governance systemsmay shape the role of and effectiveness of boundary

organizations. Moreover, little empirical research has, to our knowledge, put emphasis on

networkgovernanceacross‐scalethatisrelevantfortheprovisionofESunderclimatechange.

The objective of this paper is to address these gaps through an empirical case regarding the

provision and use of SRS in the Birriswatershed of Costa Rica.Here, both farmers and a local

electriccompanyareaffectedcorrespondentlyon‐andoff‐sitebyincreasingdegradationofSRS.

We focus on regulatory mismatches and on the identification and analysis of the role of key

actors by using policy network analysis. In the following section we address concepts and

definitionofgovernanceand its relation toecosystemmanagementandadaptation to climate

change for the case of ecosystem services. In Section 3, we describe briefly the case study

111

highlighting the drivers of SRS degradation and actors’ context of SRS provision and use. In

Section4,wepresentthemethodsofanalysisof formalpoliciesandmandatesand, finally, the

structureofpolicynetworkquestionnaireandtheanalysisofnetworkalgorithm.InSection5,key

findings are discussed alongwith existing experiences and concepts for ecosystemgovernance

andadaptationtoclimatechange.InSection6,weprovidegeneralrecommendationsforfurther

research.

5.2Concepts

5.2.1InformationandknowledgeexchangeinESgovernancestructures

Important components of effective ecosystem governance across scales are the exchange of

informationandknowledgeonpotential impactsandavailablealternatives forES conservation

(CashandMoser,2000;Adgeretal.,2005).Whileinformationistheresultofdataanalysis(e.g.

produced to understand impacts, technical solutions, etc.), knowledge results from the

interpretationandcontextualizationof information(i.e. requiresexperienceandunderstanding

ofspecificcontextswhereinformationismeanttobeapplied)(Toimu,1999).Inter‐organizational

informationnetworks,assuggestedbyInstitutionalandOrganizationalLearningtheories,provide

notonlyexchangeofinformation,knowledgeandawareness‐raisingonnewissuesbutalsohave

thepotentialtoprovideimportantinformationoncostsandbenefitsofalternativeswithahigher

persuasivenessdegreethanothersources(Brassetal.,2004).ThisisespeciallyimportantforES

contextswherealargevarietyofactors(whohavedifferentexperiences,interests,mandatesand

understandingofthefunctioningofecosystems,theirdegradationandofthepossibleresponses;

Fisher et al., 2009) interact and exchange information and knowledge in networks in order to

guide their decision‐making. The interchange of information and knowledge depends, among

other things, on politically and economically ambiguous social networks characterizing

governanceofspecificenvironmentalissues(Adger,2003;Folkeetal.,2005;Pellingetal.,2005).

Althoughdebateonthedefinitionofgovernanceisstillopen(Kjær,2004),accordingtoHuitema

etal. (2009),governance is theresultofboththeformallystatedrules (e.g.policies, laws,etc.)

and the informal interactions of actors in networks that promote collective responses to

environmentaldegradationproblems.Thus, formalmandatesand interactionsofdiverseactors

operating at different scales shape relevant opportunities and/or constraints for organizing

112

collective responses to natural resource degradation problems (Uitto, 1997; Koimann, 1999;

Hodge,2001).Buildingontheseauthors,governancethuscomprisestwointer‐relatedaspects:1)

formalrulesdefinedbylawsandofficialmandatesthatexplicitlyframeactors’legitimacyintheir

actions related to the environmental degradation problem at hand, and 2) informal policy

structuresdescribedbyinteractionsinnetworks(Piattoni,2006)thatdeterminetheirsharingof

information and knowledge across scales and policy communities23 (influenced by their

respectivepowerandinterests).

Emphasisontheinformation‐sharingmechanisminthisgovernancesystemisespeciallyrelevant

wherecomplexityanduncertaintiescharacterizingaproblemlikeESdegradation(vanNoordwijk

etal.,2004)requiretheirinformationandknowledgetobeupdatedtofacechangingconditions.

Insuchcontexts,“rigid”policiesmightproveinadequate24;rathertheestablishmentofflexible

rules,andproductionanddisseminationofinformationtoincreaseknowledgeandlearningover

time may be more appropriate (Duit and Galaz, 2008). Sharing information in networks can

promotecommonperceptionandunderstandingoftheproblem,andcapacitytoplanresponses

andmonitortheminawidernumberoforganizations(Cash,2001;vanderBruggeetal.,2007;

Pahl‐Wostletal.,2009).

5.2.2MultiplescalesandregulatorymismatchesforSRS

LandusedecisionsinfluencingboththeprovisionanduseofESaretypicallytakingplace inside

institutionalstructuresthatspanacrossscales(RotmansandRothman,2003).Forourcase,the

life span of downstream hydropower dams depends on the quantity of sediment transported

from uplands (Southgate andMacke, 1989; Guo et al., 2000) while the impacts of erosion on

farmersarevalued inupstreamareas (Pimentel etal., 1995). Thepotential jointgainsamong

these two scales depend on whether there is coordination and information exchange among

differenttypesofactorsoperatingatdifferentscalesandembeddedinlargernetworks.

23 Policy communities are defined by actors sharing experience, common specialist language. A subset of actors in larger

networkssodefinedas“policycommunity”sharesacommonidentityintermsoflanguagesandday‐to‐daypolicy‐making.Forexamples,scientistshavelimitedpresenceintheday‐to‐daypolicy‐makingactivitiesoftheregulators’communitiesand

vice‐versa(HowlettandRamesh,1998).

24RigidpoliciesarereferredtobyDuitandGalaz(2008)asthosethatcanbeelaboratedbasedonhigh scientificcertaintyandclarityonthebehaviourtobe(suchasprohibitiontosmokeinpublicplaces),thenclear‐cutlawsareeffective.

113

Scalesarethusasocialconstructandtheirdefinitiondependsontheperspectivetaken(Cashand

Moser, 2000). Along with these authors, we understand that “scale refers to a specific

geographically or temporally bounded level atwhich a particular phenomena is recognizable”.

Morespecifically,erosionisrecognizableatthefarmlevelwherefarmersmightperceiveitasa

problemaffecting their soil fertility, at thewatershed scalewheredownstreamusersmightbe

affectedbyupstreamactivitiesand, finally,at thenationalscalewhereagriculturalpoliciesand

scientificorganizationsareinterestedinpromotingsoilconservationandagriculturalproduction

potential. Aswe have seen, the degradation of SRS as a social‐ecological problemdepends on

components that range from the very local level (e.g. topography and soil; farm land use

decisions)tothewatershed(e.g.microclimaticconditions;presenceofconservationprograms),

to the national (e.g. distribution of precipitation; laws and incentives), and finally to the

international (e.g. climate change and variability; markets; and, international agreements and

funding) scale. Access to adequate resources such as information and incentives, needed to

supportstakeholders’responsestoESdegradationatagivenscale,isconditionedbycross‐scale

institutionalgapsandmismatches25(CashandMoser,2000).

5.2.3BoundaryorganizationsforESgovernancenetworks

Structural interactions of actors in networks has received increasing attention in social studies

(Borgatti, 2009) and specifically in studies of environmental degradation problems and

adaptation to climate change (Adger, 2003; Brooke, 2002; Jost and Jacob, 2004). Actors’

information exchange in networks is strategically important to effectively plan adaptive

responses (DuitandGalaz, 2008).BoundaryOrganizations (BOs)arekeyactors in information‐

sharing networks, capable of bridging information across scales and knowledge systems.

Boundary organizations provide spaces for changes in cultural aspects by facilitating mutual

understanding of preferences and meanings among different epistemic communities. Indeed,

they promote the creation of boundary packages, such as collaborative partnerships between

farmers, scientists, and policy makers PES schemes (Guston, 2001). Additionally, BOs also

25Forexample,thescaleofinformationproducedbyanactoratnationallevelsuchasacountry‐levelclimatechangeimpactstudiesmight require, inorder toguideactiononES,downscalingof information toa specific siteand communicationof

uncertaintiestonon‐expertaudience.Similarly,nationalactorssuchasregulatorswouldimprovetheiractionsifbuildingonlocalknowledge/experienceonsoilerosion(e.g.constraintsandinnovativesolutioninapplyingsoilconservationpractices).

114

contribute to the stability of multi‐stakeholder processes (involving accountability and

participation) while maintaining information‐sharing mechanisms between science and policy

and supporting both the building and performance of partnerships across scales (Cash 2001;

Guston,2001;Cashetal.2003).

Structural analysis of information‐sharing networks can allow identification of key bridging

organizations(Burt,2005).BOscan,thus,alsobedefinedbystructuralnetworkpropertiessuch

asthenumberofinformationlinksacrossscalesanddifferentpolicycommunities,heterogeneity

oftheactorsconnectedandthedecentralizationlevelofinformationthroughitsnodes(Weible,

2008).ThisstructuraldefinitionhighlightsthepotentialroleofBOstocontributetoanetwork’s

capacitytoi)definewhatandwherearetheESdegradationpriorities,ii)whicharetheavailable

solutions and, finally, iii) how institutional mechanisms could be shaped to implement and

monitor responses (Carlsson, 2008). Structural analytic methods use Betweenness Centrality

(Freeman, 1977) as a measure of actors’ capacity to mediate and influence relations among

actors(BodinandCrona,2009)andgeneratecontagionofideas(Borgatti,2009).

5.3.ContextfortheBirriscasestudy

The Birris sub‐watershed, part of the upper Reventazonwatershed (1500Km2), has an area of

almost five thousands hectares and its population density of 161 inhabitants per square

kilometreisabovethenationalaverage(INEC,2002).SRSdegradationisdeterminedbyexposure

to extremeprecipitation, by the area’s high sloping topography and by land use conflicts. Its

steep landssuffer serious soil erosiondue to intensiveusebyhorticulturalistsanddairy‐cattle,

whichcomprisethemajoreconomicactivity. Itsvulnerabilitytoclimatechange ischaracterized

by observed increases in extremeprecipitation events over the last forty years (Aguilar et al.,

2005)andbytheprojectedclimatechangescenariosfortheregion(Magrinetal.,2007).Thehigh

fragmentation of agricultural family plots and the high market orientation of its production

represent a challenge for existing conservation incentive mechanisms such as PES which are

challenged by high opportunity costs to local farmers (Pagiola, 2008). In this context, the

downstreamhydropowerdamsoftheAdministrativeBoardofElectricServiceofCartago(JASEC)

and of theNational Electric Institute (ICE) suffer the off‐site effects of upland erosion. Several

millionsUS$arespenteachyeartoflushanddredgesedimentsoutofthesedams.Thesefigures

115

are of national interest given that i) the Reventazon watershed is the most important in the

countryforhydropowerproduction;ii)thathydropowerrepresentsmorethantheeightypercent

ofenergysourceinCostaRica;iii)thatcurrentpoliticalgoalsaimatincreasinguseofhydropower

asanationalstrategy(ICE,2007).

5.4Methods

When considering environmental governance, we focus on the analysis of formally stated

objectivesandmandates(e.g.asdefinedbynational laws,decreesandofficialstatements)and

actors’interchangeofinformationwithothers.Definingmanagementofsoilregulationservices

inthecontextofwatershedsrequiresconsiderationofthemultipleactors fromdifferentpolicy

communities (Van Noordwijk et al., 2007). Indeed, governance for provision of ecosystem

services in agricultural landscapes can be described by the interactions of actors focusing on

socialandecologicalsystems(Hodge,2001;Ostrom,2007).Drawingfromthepluralisticapproach

(Cowie, 200526) and the trialoguemodel forwater governance27 (Turton et al., 2007), and the

specificSRSdecision‐makinginterestedactors(Fisheretal.,2009),ourSRSenvironmentalpolicy

systeminvolvesfourpolicycommunities:

1. Directusersofoff‐sitebenefitsofSRS(e.g.hydropowerdamsdownstream);

2. Direct users of on‐site benefits of SRS (also influencing provision off‐site) in upstream

areas(i.e.farmers,theirassociations,civilsocietyinterestedinupstreamconservation);

3. Scientistsprovidingunderstanding,knowledgeontechnicalaspectsofSRSprovision(role

of forests, soil, agriculture and climate on soil erosion) and those involved in

disseminationofscience;and,

4. Regulators providing policy guidelines relevant for the management of watershed and

theirSRS.

26 Cowie (2007) indicate the potential stakeholders to be involved in the design of collective actions for watershedmanagementplanningsuchasrules‐makers(e.g.governmentagencies),actorsdirectlyinvolved inbenefitsand/orcostsofdecisions(i.e.providersandusersofSRS)andpartieswithtechnicalknowledge(e.g.scientistsandtechnicalagencies).27Theseauthorsidentifygovernment(rulemakers),scientistsandcivilsociety(actorsdirectlyandindirectlyaffectedbyESmanagementdecisions)asthreeimportantpillarsofwatergovernance.

116

Defining the network boundary represents the first step in the application of social network

methods(Knoke,1990;WassermannandFaust,1994).Weidentifiedactorsbymeansofatwo‐

stepcomplementaryprocedure,consultingkeydocumentsandinformants(Jost,2004).First,we

identifiedlawsandofficialmandatesrelevantforeachpolicycommunity(PenkerandWytrzens,

2008).Whereorganizationswereexplicitlymentionedinthelaws,theywereincludedinthelist

of actors. Secondly, we complemented this initial list using reputational snowball techniques

(Farquharson, 2005) asking key informants who represented each of our four policy

communities) at different scales to list organizations that they considered to be influential in

issues such as soil conservation, climate studies, land planning, watershed conservation, and

hydropowerproduction.Then,weexplicitlydefinedthoseorganizations(associations,ministries,

privatecompaniesandNGOs)thatactivelycontributetosettingpoliticalagendas,activities,and

proposals,orthatparticipateintheimplementationofpoliciesandactionsinoneofthoseareas

relevant for SRS provision and use. Because individual respondents at one scale within an

organization may not acknowledge all interactions of all departments of his organization,

(Brinkerhoff 1996), we consider sub‐units of such organizations (e.g. departments) as actors

themselves.Thus,we identifiedtheactors/organizationsconcernedwithformulation,advocacy

and selection of alternatives to respond to SRS degradation problems (Knoke, 1990). For the

analysis of formal mandates and objectives, for each identified organization we reviewed

published documents and characterized their relevant tasks for designing and implementing

actions to improve SRS. During the network questionnaire meetings we asked open‐ended

questionsto intervieweesonwhataspectstheyperceivedasabarrierorasanopportunity for

improving their support to SRS provision (McDaniels et al., 2006). These interviews were

transcribed and used to cross‐check and complement our secondary data on laws and official

mandatestoanalyseformalpolicycontext.

5.4.2Networkanalysis

Reflecting theworkofBorgatti (2009),wehave considered fourbasicaspectskey toanalyzing

multi‐actor interactionsusing socialnetwork theory: theoretical framework, research question,

relationsandstructure.OurtheoreticalframeworkrestsontheoriessuchasInstitutionalTheory

(DiMaggioandPowell,1983),OrganizationalLearningTheory(Levitt&March,1988)andtheory

ofAdaptiveGovernance (DuitandGalaz, 2008) all ofwhichemphasize the roleof information

117

transmissionamongactors’networkstopromotelearninganddesignofcollectiveresponses(see

Section 2). Our research question aims to identify key bridging organizations in order to

subsequently contrast their boundary positions with evidence on their current and potential

ability to contribute to networks’ adaptive capacity. Relations refer to the content of the

connection among couples of actors. In our case, this would include information or resource

flows connecting two ormore actors for its relevance to learning (Duit and Galaz, 2008). The

concept of structure refers to the position of ‘nodes’ (i.e. organizations)with respect to other

actorswhenagivenrelationisconsidered.Weemployawidelyusedconcepttostudystructural

positioninnetworks:‘betweenesscentrality’(BC)(Freeman,1977),whichmeasuresthebridging

capacityofanactortoconnectotherpairsofactors.

5.4.2.1Buildingthequestionnairefornetworkanalysis

Weanalysedthreetypesofrelationsbasedonpreviousresearch(Knoke,1990;Cashetal.,2003;

RaabandKenis,2006):

1) Position in inter‐organizational information‐flow networks as a proxy of their bridging

functions between local watershed level and national policy and science‐making; this is

calculatedbyananalysisofapolicynetworkmatrix(seebelow);

2) Perceived influence as a proxy of the persuasive power of an actor given by how he is

perceivedbyothers;

3)Perceivedcompetenceasaproxyoftheperceivedcredibilityof informationproducedbyan

actor by other actors which can influence how effective is this information in promoting

responses.

BuildingonKnoke(1990)andWeibleandSabatier(2005),wepresentedalistoforganizationsto

everyrespondentandaskedthemtousethecorrespondingscalestoindicatehowtheyrelateto

otherorganizations,namely:

1) Please check the corresponding box for those organizations from which you receive

informationusefulforyourdutiesandfunctionsinyourownorganization(i.e.binaryvariable)

2) Please check the corresponding box for those organizations to which you give information

producedbyyourownorganization(i.e.binaryvariable)

118

3)Onascalefromonetofourpleaseratehowinfluentialyouperceiveeachoftheorganizations

inthe list (wherenocheck isnoopinion,1 isvery low influenceand4 is thehighestperceived

influence)

4)Onascalefromonetofourpleaseratehowcompetentyouperceiveeachoftheorganizations

in the list is (where no check is no opinion, 1 is very low competence and 4 is the highest

perceivedcompetence)

5.4.2.2Matrixanalysisofnetworkdata

Networkdatarecompilationwasusedtocreatesymmetricalmatrices(i.e.adjacencymatrix)with

nodes(i.e.organizations)inthefirstcolumnsandinthetoprow,filledwiththevaluesassigned

byeach respondent (on the row)abouthis relation to theothers (i.e. listed in the topofeach

column) (HannemannandRiddle, 2005). Forour study,each relationwasanalyzed through its

specificmatrix, although combinations of relations can be visualized in a single graph. To run

analysisofthematricesweusedUCINET(BorgattiandEverett,2002).Tovisualizeresultsweused

its complementary software NetDraw which allows representation of nodes, relation, and

networkparameters(Brandesetal.,1999).Severalalgorithmsfornetworkanalysisareavailable

intheUCINET‐NetDrawpackage.Forour informationnetwork,weanalyzedparameters largely

usedinpolicynetworkanalysis(Knoke,1990).Wefirstconfirmedinformationflowsbychecking

whetherstated“giveto”informationflowisconfirmedbythereceiver.

Then,weusednetworkalgorithmsforstructuralanalysisofoverallnetworkcharacteristicssuch

as density (defined by the total number of existing ties out of all possible ties) and actors’

Betweenness Centrality ( )BC . Density, defined as ameasure of group cohesion, takes a value

between0(noties)and1(allactorsinter‐linkedtoeachother)istheaverageofthestandardized

actordegreeindexaswellasafractionofthenetworktiesforagivenrelation(Wassermanand

Faust,1994:181).BetweenessCentrality ( )BC measuresthebridgingpositionofanactorandis

determined by an actor’s position with respect to two others (i.e. the minimum required for

definingbridgingpositions)(Knoke,1990;Brandesetal.,2003).Thispositionmeasureofanactor

isformallydefinedbyWassermanandFaust(1994:190)as:

( )( )

jk i

B i

j k jk

g nC n

g<

=!

119

Where ( )B iC n is the Betweeness Centrality of actor i , jkg represents the number of geodesics

linkingactor j withactor k .Then ( )jk ig n is thenumberofgeodesicsbetweenthesetwoactors

thatpassesthroughactor i .Thisalgorithmassumesthatlinks(i.e.informationpaths)haveequal

weight (i.e. probability to be chosen), and that communicationswill travel along the shortest

route(regardlessoftheactorsalongtheroute).Thisindexisthenasumofprobabilitieswitha

minimumof0(iftheactorfallsonnogeodesics)andmaximumgivenbythenumberofpairsof

actors not includingin reached when i falls on all geodesics. Finally, we analyse the average

values of perceived influence ( )Pi andperceived competence ( )Pc that are attributed to each

actorbytheothersinthenetwork.

120

5.6Results

5.6.1Networkboundary,formalpoliciesandmandatesoforganizationsacrossscales

Basedonkeyinformants,analysisofpolicydocumentsandface‐to‐faceinterviewsweidentified

theboundaryofthenetwork(i.e.theconstellationofactors)(Table1).

Table 1: Actors and their scale scope defined by the network boundary for improving SRS in Birris watershed (in Appendix 1 see extended version of organizations’ acronyms, formal mandates).

Scale Policyarea

Regulation Demand Supply Science

Organizations1

Internationaltonational

CRRH,IICA,IUCN

National ARESEP,Agua‐MINAE,

FONAFIFO,CIA

IPN,IMN,INTA,CIA‐UCR,ENCC,PROGAI‐UCR,CADETI

Nationaltolocal

MAG,SINAC‐MINAE,SENARA‐

MAG

ICE,ICE‐UMCRE

IDA,CNP,IMAS CNE

Regional JASEC COMCURE ASA‐Pacaya

Local Municipality Producersassociations(ADICO,ADAPEX,COOPEBAIRES

1Extendedorganizations’namesareinAnnex1.

At the national scale, actors such as IICA and CRRH disseminate information on climate or

agriculture‐related agreements28 that influence benefits and costs of policies dealingwith ES‐

related land uses in the country by providing inputs to design laws, research programs and

mechanismstopromoteSRSconservation.Similarly,largecivilsocietyorganizationssuchasIUCN

organize and participate in national and international initiatives to raise awareness on climate

changeandtheroleofecosystems inadaptationpolicies.Wealsofindorganizationsregulating

and providing incentives for water resources and watershed management in general such as

28UnitedNationFrameworkConventiononClimateChangeor theUS‐CentralAmericaAgreementonFree Trade (CAFTA;withastrongagriculturalcomponent).

121

FONAFIFO,ARESEP andAgua‐MINAE. Forexample, these typesof organizationsestablish rules

that affect the economic efficiency of SRS provision to hydropower production. In the science

policyareawehaveorganizationsdedicatedtoproductionofscientificstudiesonclimatechange,

itsexpectedandobservedimpactsandonbestpracticesforSRSsuchasIMN,CIA‐UCR,CADETI

andINTA‐MAG.Scientificinformationismainstreamedinnationalpoliciesthroughgovernmental

initiatives such as IPN and ENCC also promoting strategic partnerships to initiate actions on

mitigationandadaptation.Otherorganizationsusescientificinformationtodefinepriorityareas

andregulatehowlandusemanagementandchangeshouldcomplywithtechnicalstandardsto

promoteSRSprovisionsuchasCNE,MAGandSENARA.Wealsofindgovernmentagenciesthat

supportprogramsto improvefarmers’well‐beingandreducetheirvulnerabilitytomarketsand

toclimateextremesthroughtraining,disseminationoftechnical informationandbypromoting

partnerships of farmers’ associations thus improving their landmanagement practices such as

IMAS, CNP and IDA. On the demand side of SRS, hydropower company ICEhas interests and

promoteswatershedmanagementinitsprioritywatershedssuchasReventazonthuspromoting

partnershipswithotheractorsalsointerestedincontrollingerosion.

At the regional watershed scale, intermediation between national directives and local

implementation is key to provision feed‐back information in both directions. For example,

nationalpoliticaldirectivesstrivetoempowerwatershedcommitteessuchCOMCUREtoestablish

participatoryanddecentralizedactionstoconserveprioritycatchmentsalso inpartnershipwith

anotherhydropowercompanydemandingSRSasJASEC.Atamorelocalscale,undertherecent

decentralizationeffortsofthenationalgovernment,thelocalmunicipality(MUNI)ismandatedto

regulate landmanagement. In upstream locations, farmers associations, representing small to

medium producers (ADICO, ADAPEXand COOPEBAIRES), share common goals of promoting

productivecapacitiesandmarketingofagriculturalproductsaswellasinformationexchangeon

alternativetechnologiesrelevantforSRS.

5.6.2Structuralanalysisofthegovernancenetwork

ThenetworkrepresentationinFigure1highlightssomeimportantactorsastotheirinformation‐

bridgingpotential.Values inTable2givemorequantitative informationon thenetworkvalues

associated with each actor. Considering that density of ties equal to one represents full

connectionsofallnodes,theoveralldensityofournetworkislow( 0.12D = )withonly214ties

122

present (i.e.outof1722possibleties).Our results fromtheanalysisof BC highlightkeynodes

thatbridgeinformationacrossscalesandpolicyareas29.Althoughthereisarelativecontinuumin

therangeofvalues max min( 52.16; 0.62)BC BC= = , thefirst fiveorganizations30 (IMNethnicunit,

ICE environment, CNE, ASA‐Pacaya and regional SINAC) in Table 2 show high BC values

highlightingtheirbridgingroleforinformationexchangeacrossscalesandpolicyareas.

Figure 1: Structure of the information network showing the actors (node) in quadrants by scale (x) and policy areas (y), the confirmed information flows (arrowed links) and its Betweenness Centrality (size).

29We tried an analysis of only reciprocal ties where actor x and y interchange information in both direction as tried byHannemanRA,RiddleM.(2005).(Hanneman,R.,R.A.,Riddle,M.,2005.Introductiontosocialnetworkmethods.Riverside,California , available at: http://faculty.ucr.edu/hanneman/.)When running this analysis, IDA‐regional is themost centralactorofthewholenetworkandtherestoforganizations’centralitiesaremoreevenlydistributed.However,fromthispointon,wewilldiscusstheresultsofourconfirmed“giveinformationto”matrixthatallowsidentificationofdifferenthigh‐Bcvaluesintheinformationflowmechanismatdifferentscales.30 Considering the continuum of values, in correspondence of the fifth organization there is the highest value‐jump/discontinuity(i.e.6points)tothesixth.

123

The vector measuring perceived influence max min( 3.6; 1.6; 2.9; 3.1)average medianPi Pi Pi Pi= = = =

correlates with the vector measuring Betweenness centrality

( 0.407; 0.05)PearsonR p= < suggesting that bridging positions in the network are associated to

persuasive power. On the other side, Perceived Competence

. .( 2.6; 0.3; 2.7)Mean S D Median

Pc Pc Pc= = = isnotcorrelatedtovaluesofBetweennessCentralitybut

issignificantlyandpositivelycorrelatedtoPerceived Influence ( 0.54; 0.0001)PearsonR p= < (Table

2).

124

Table2:BetweennessCentrality(BC)andPerceivedInfluenceandPerceivedCompetencevaluesin

informationnetwork(organizationslistedfromlargerBCvaluestosmaller).

Perceivedinfluence(Pi) PerceivedCompetence(Pc)

Ties Betweennesscentrality(BC)

Organizations n Pi S.D. n Pc S.E. In Out BCIMNTécnicos 17 3.3 0.9 17 2.8 1.0 11 7 52.2ICEAmbiente 15 3.2 0.8 16 2.7 0.8 5 15 47.4

CNE 26 3.3 0.9 24 2.8 0.9 8 5 39.2ASAPACAYAS 6 3.3 1.2 6 2.8 1.0 2 9 38.0ICEElectricidad 12 3.4 0.8 11 3.1 0.7 2 16 37.1SINACCentral 18 3.4 0.7 17 2.5 0.9 6 10 31.4PROGAI‐UCR 17 2.8 1.0 17 2.7 1.0 9 5 31.3IMNDirección 21 3.4 0.9 21 2.8 1.0 9 4 31.0MAGRegional 21 3.6 0.8 19 3.1 0.9 4 7 29.8ACCVC‐Cartago 19 2.7 1.0 19 2.5 0.9 5 8 28.2

ENCC 21 3.1 1.0 20 2.5 1.1 9 5 27.4INTAPlanton 12 2.7 1.2 10 2.4 0.7 7 6 27.2AGUASMINAE 27 3.4 0.8 27 2.7 0.7 9 3 24.4

INTAST 14 3.1 1.0 13 2.7 1.0 3 9 23.2IDANacional 18 3.1 0.8 18 2.3 0.6 9 3 20.1CNPRegional 14 2.7 0.8 14 2.1 1.0 6 3 18.1

SENARARegional 23 3.2 0.8 23 2.5 0.8 5 7 17.3IDARegional 17 2.9 1.1 15 2.7 1.0 2 6 16.6

JASEC‐Proyectos 17 3.2 1.1 15 2.9 1.1 6 6 16.6FONAFIFO 26 3.4 0.8 25 3.2 0.9 9 2 15.5

IPN 15 2.8 0.8 15 2.1 1.0 9 2 13.9CNP‐Subregional 12 2.1 1.1 11 1.9 0.9 2 4 13.9

MUNI 15 3.1 1.1 13 2.3 0.8 5 6 13.8JASEC‐Gerencia 13 3.2 0.8 12 2.8 0.9 5 5 13.5

UMCRE 16 3.3 0.9 16 3.2 1.0 4 6 13.3CRRH 12 3.3 1.0 12 2.9 1.2 6 2 12.9

COMCURE 20 3.3 0.9 18 2.9 0.9 10 1 10.0ICECENPE 13 3.3 1.0 13 3.0 1.1 5 5 7.6

IMAS‐Regional 12 2.8 1.2 12 2.8 0.9 3 2 7.1ICEPISA 12 3.1 1.1 12 2.8 1.1 3 6 6.8

ARESEPEnergía 16 3.3 1.0 15 2.5 1.1 7 1 6.8ADAPEX 9 2.2 1.3 9 2.9 1.2 5 0 6.0CIA 16 2.6 1.2 16 2.3 0.9 3 5 5.8

JASEC‐Producción 7 2.6 1.1 7 2.6 1.0 1 7 5.7IICA 14 2.6 0.9 14 2.5 0.9 6 1 5.5

ADICO 6 1.7 1.0 6 2.5 1.1 1 4 5.1COOPEBAIRES 7 2.3 1.1 7 2.4 1.0 4 1 4.6SINAC‐ACCVC 18 3.1 0.8 18 2.4 0.9 3 5 3.8

CADETI 6 3.5 0.6 6 2.5 1.2 0 6 3.7ICEProduccion 12 3.5 0.8 12 3.3 0.9 3 2 2.5

CIA‐UCR 8 2.1 1.1 8 2.3 1.2 2 3 2.0UICN 5 2.6 1.1 5 2.8 1.1 1 4 0.6

Atthelocallevel,actorssuchasprivateprovidersofagriculturalinputsandfarmers’associations

(COOPEBAIRES,ADICO,ADAPEX)showverysmall BC values.Theirmaininformationsourcesare

agenciespromotingfarmers’productiveactivities(ASA‐Pacaya,regionalCNPandMAG,andCNE).

125

Watershed level isascalewhereorganizationscanbe importantto informationexchangefrom

local to national scales and across different policy areas. Departments of JASEC and ICE,

exchangeinformationamongthemselvesandwithotherorganizationsattheregionallevelthat

focusonSRSscienceresearchanddissemination,suchasINTA‐Planton,MAG‐regionalandASA‐

Pacaya,allofwhomshowhigh BC andaperceivedcompetenceandinfluencevalue( 2.8Pc = ;

3.3Pi = . These values are among the highest in the network (i.e. in the percentile 75).

COMCUREshowsalow BC value.Atthenationalscale,the largestBetweenessCentralityvalue

of the IMN ‐technical department is consistent with its mandate to provide information on

climatechangetoawiderangeofactorsindifferentpolicyareas(e.g.technical informationon

extreme events that might affect hydropower production and/or agricultural producers).

Moreover, in linewith itsmandate to coordinate the National Communication toUNFCCC on

bothmitigation and adaptation, IMN recompiles information frommany sectors to synthesize

prioritiesofdifferentsectors.CNE,asanationalorganizationwithlocalbranches(e.g.community

emergencycommittees),bridgesinformationamongusersofSRSatnationalandwatershedlevel

(e.g. ICE‐PisaandJASECcorrespondently),scientificorganizations(i.e.national INTA researching

on soil conservation alternatives) and local committees dealing with landscape management.

Regulatory agencies such as MAG, MINAE, SENARA and institutionalized multi‐stakeholder

committees suchasENCC, IPN,andCADETI have relativelymediumto low BC valuespossibly

due to their recent formationand/ormarginal role in information sharingmechanisms. Finally,

actors at the international‐national interface such as IICA, CRRH and IUCN show low values

possiblyduetothefocusofourlocalcasestudy31.

5.7Discussion

Our results show that beyond the institutional and formal policy framework, organizations

interactandsupportinitiativesacrossscales,buildinginformalpartnershipsandlinkagestocope

withtheirSRSdegradationproblems.Ourformalanalysisofnetworkshasshownthatthereare

keynodes/organizationsthathelpprovideinformationacrossscalesandpolicyareasandwhose

importantroleshouldbere‐evaluatedbynationalpolicies.

31Wecouldhaveexpandedtheboundaryofthenetworktotheinternationalarenabutpotentiallycouldhavelostrelevantdetailsonthenationalnetworksthatinfluencepolicy‐makingforourlocalcasestudy.

126

Someauthorsinrecentarticleshaveoutlinedtheroleofnationalpoliciestoprovideanenabling

institutionalenvironmentandappropriateresourcesforcross‐scalegovernanceofenvironmental

degradationproblems(Adgeretal.,2005;DuitandGalaz,2008).TheintendedgoaloftheCosta

Rican national policy on soil resources (Law 7779) is to “protect, conserve and improve soil

resources in an integrated and sustainable management of other natural resources and by

promotingadequateenvironmentalplanning”, identifying theMinistryofAgriculture (MAG)as

theimplementingagency.Thelawsetstheframeworktoprovideincentives,technicalassistance,

and formation of management committees to foster soil and water conservation. More than

hundredlawsinthecountrydealwithwater‐relatedissues,hinderingtheidentificationofclear

administrative responsibilities for protecting water related ecosystem services such as SRS

(Segura,2004;Mirandaetal., 2006). Inspiteofthis institutionalcontext,ourresultsshowthat

severalactors,followingtheirmandatesand/orinterestshavepromotedcollaborativeeffortsto

buildcapacity,knowledgeandactions(Mirandaetal.,2006)atdifferentscales.

Forexample,IMNalongwithregionalandglobalinterests(PACADIRHandGEF),hasimplemented

projectsonadaptation to climatechange inareas characterizedbyvulnerablewater resources

which have involved many national (including ICE andMINAE) and local organizations. These

results support findings by previous research on inter‐organization networks for biodiversity

conservation and adaptation to climate change (Brooke, 2002). National organizations such as

CNE, IMN, MINAE have conformed to the recommendations of the National Consultative

Commission for Climate Change (2003) to promote adaptation responses in the country.

Similarly,MINAEtogetherwithMAGandCADETIhavebeenpromoting,since2004,aninitiative

topreventlanddegradationthroughcollaborationwithcivilsociety,andlocalinitiativesforland

planningandeconomicdevelopment.

At the local level, farmers, affected by on‐site effects of SRS degradation in upstream areas

searchfor informationontechnologicalalternativesthroughtheirnetworksofprivateagencies

dealingwithagriculturalinputs32and/orthroughinstitutionalchannelsofgovernmentextension

officessuchasASA‐PacayaorMAG.Theseorganizationstransfertechnical/scientificknowledge

onsoilconservationalternativesproducedbynational researchorganizationssuchas INTAand

32Inthecaseofprivateextensionservicesdealingwithagriculturalinputstheremightbeaaconflictofinterestsforthebiastowardsmarketingoftheirownproductstofarmerswhichnotnecessarilyareenvironmentallyfriendly.

127

CIA‐UCR.Similarly,actorsaffectedbytheoff‐siteeffectsofSRSdegradationsuchasJASECandICE

are involved in initiatives to copewithSRSdegradation.These typesofactors implementboth

private corporate responses (Smit et al., 2000) and use institutionalized structures to build

alliances for planning and implementing adaptation responses. For example, JASEC and ICE,

directly bearing the costs of sediments removal from their reservoirs, have adopted corporate

restructuringofdepartmentsreflectingtheneedtorespondtoSRSdegradationcosts.TheJASEC

managementboardhasadoptedaformalresolutiontoinstitutionalizeawatershedmanagement

unit to channel its resources to concrete conservation actions. Similarly, ICE has placed its

formerly separate watershed management unit33 ICE‐UMCRE under the production units to

monitor and cost watershed conservation actions against evidence of benefits in reduction of

operation costs (i.e. sediment removal). As part of these responses, these organizations

promoted collaborations with the ASA‐Pacaya extension office to support dissemination of

innovativesoilconservationpractices.

Thecapacityofthiscross‐scalegovernancesystemtorespondtoSRSdegradationisinfluencedby

distribution and access to funding (e.g. to sustainmonitoring and learningmechanisms across

scales;Brooke, 2002)and information (e.g.onalternatives to respond to changingconditions).

Distribution and access to these resources is influenced by institutionalmismatches (Cash and

Moser,2000).Twoexampleswillhelphighlightinghowinstitutionalmismatchesandinformation

gapscanhinderactors’capacitytocopewithESdegradationproblem.Wethenshow,through

structuralanalysisofinformation‐sharingnetwork,thatBOcanbeanimportantgovernanceasset

tohelpovercomingthesegapsandmismatches.

Thefirstexamplereferstoaccesstofundingforwatershedconservationinitiatives.ARESEP,the

regulatorof tariffs forpublic services, requires sound technicalevidence forassigning rights to

access financial incentives tohydropower investing inwatershedconservation. Inotherwords,

the hydropower facility must provide sound scientific evidence approved byMINAE, that the

investments inwatershedconservationactivitiesarenecessary for the sustainableprovisionof

electric service34. The large uncertainties characterizing SRS provision under climate change

33Thisunitfocusedonbroadintegratedwatershedmanagementactivitiesnotspecificallytighttoproductionofhydropower.34Inthisway,themarginal investmentsmadetopromotewater‐relatedecosystemservicessuchasSRSinthewatershedscanbeconsideredascorrespondingtothemarginalcostofproducingtheserviceandassuchcanbedetractedfromthetaxamountdue.

128

representaseriouschallenge.Inthisrespect,ICE isautonomouslypromotingthe installationof

monitoringstationstomeasuresedimentsfromprioritywatershedswhereconservationactivities

arecurrentlybeingsupported(i.e.toincreaseevidence‐basedinclusionofSRSconservationcosts

inelectrictariffs).Generating informationonsoilconservationactivitiesand itseffectsrequires

ICEtocollaborateandexchangeinformationovertimewithlocalpartnerswhointermediatewith

farmers.

Asecondexampleofscalemismatchisof“assessmentandmanagement”type(CashandMoser,

2000) and is related to the contrast between the activities, defined by national actors, to be

funded for watershed protection (based on the FONAFIFO list of eligible activities) and the

realitiesoflandusesinwatershedswhereJASECandICEareoperatingthatrequireupdatingof

information.EligibleactivitiesundertheFONAFIFOschemeincludeonlytree‐basedlandusesasa

way to protect the provision of water services, used for the calculation of the environmental

tariff(tointernalizetheSRSconservationcosts;Ponce,2006)andisbasedonuncertainscientific

evidence (Locatelli andVignola, 2009).Moreover,emphasison treesexcludes priorityareasof

watershed such as Birris. Here, providing support to technical information agencies (e.g.

improving soil conservation practices) would be amore acceptable alternative to current PES

incentives(Vignolaetal.,2009)whichcannotcompensateforthelargeopportunitycostsfaced

bysmall‐holdershorticulturalists.

Thesemismatchessuggestthatpolicyobjectivestoachievesustainableprovisionofservicesare

challenged by inherent uncertainties in estimating costs and benefits of SRS conservation

responses.Inthiscontext,information‐sharingmechanismscanimprovecapacityofinstitutions

to target investments (i.e. where and what actions) and monitor their effects over time. Our

structuralanalysishighlightsthatsomekeyorganizationsatthenational levelsuchas IMNand

CNE operate within networks for coping with climate change vulnerability and biodiversity

conservation issues in Costa Rica (Brooke, 2002). Although, SINAC, UCR, MAG and other

organizationsappearatthetopofthelist,thehigh BC valuesoftheagriculturalextensionoffice

ASA‐Pacaya deserve discussion.ASA‐Pacaya hasenabled connections between local direct SRS

stakeholders (e.g. providers and users) to others across scales and policy communities. It is

crucialtonotethatattheintermediatescaleofthewatershed,bridgingpositionsareespecially

important. While national policies have designed institutional spaces such as COMCURE to

129

mediateacrossscalesandsectors,theroleofadecentralizedwatershedmanagementinstitution

in information sharing is not yet fully acknowledged by other organizations in the network35

(possiblydueto limitedexperienceandresourceendowment;Diazetal.,2009).Watershedsoil

conservationinitiativescurrentlyinplace,however,suggestthatothermediatorsarefacilitating

relevant partnerships. The bridging position ofASA‐Pacaya, the only non‐national organization

amongthetopfive BC values,resultsfromitsinvolvementininterchangeofinformationacross

scalesandsectors(suchasthoseforSRSconservation).ASA‐PacayaisinvolvedinSRSinitiatives

withactors fromdifferentknowledgeareasofSRSpolicysuchasJASEC, ICE, INTAandCIA‐UCR

reflecting its boundary role. Indeed, this result supports the findings of Cash (2001), which

outlinedthekeyroleofagriculturalextensionorganizationsasboundaryorganizationsspanning

knowledge systems and ofMahanty (2002)who showed that intermediary positions (between

thestateandvillagers in India)arecrucial in facilitatingconservation interventionsforbuilding

trustandthedesignoflegitimatesolutions.

Multiple knowledge systems are required to address the complexities and uncertainties

associatedwithestimatingclimatechangeimpactsonwatershedSRSandthedesignofresponses

(Cashetal.,2003;Hulme,2005;Pellingetal.,2005;DuitandGalaz,2008).ThehighBCvaluesof

ASA‐Pacaya reveal its knowledge‐sharing potential36.ASA‐Pacaya plays an important boundary

role while i) communicating research across scales, ii) translating messages into appropriate

languagesfordifferentsystemsofknowledge(i.e.policyareasorcommunitiesofpractice)andiii)

mediating among the different interests characterizing them. To enhance this role, boundary

objects (e.g. reports, forecasts, models) produced by boundary organizations should be

accountabletobothscienceandpolicycommunitiesandthroughdifferentscalesoforganizations

(Tuinstraetal.,1999;Guston,2001).Anexamplefromourresearchmighthelphighlightthisrole.

ICE needs information on climate extremes events provided by IMN and on the effect of

conservationpracticesinplaceinthewatershedonerosionprovidedbyINTAorCIA‐UCR.These

actors promote and implement soil conservation research projects in collaboration with ASA‐

Pacaya that directlyworkswith farmers. The activities ofASA‐Pacaya include the provision of

35Asshownbythelowcitationininformationexchangebyotheractors.36 A similar use ofBC can be found in the analysis of Leydesdorff (2007) on the knowledge‐sharing potential of scientificjournals where high BC values indicated journals with “strong trandisciplinary contents” (i.e. able to transmit, translateknowledgefromdifferentacademicfields).

130

technicalinformationonsoilconservationalternativestolocalfarmers,supportofnationaland

international research initiatives (e.g. INTA‐INIA’s Planton Pacaya soil conservation project; or

IUCN‐CATIEproject37)focusingonsoilerosioncontrolandclimatechangeimpacts.Theresultsof

suchresearcharepresentedinnationalforaanddisseminatedtorelevantnationalcommittees.

SimilartothecaseoutlinedbyCash(2001),ASA‐Pacayasupportseffortstodefinewaterandsoil

management as a multilevel problem connecting national academic research, watershed

committees,providers(i.e.thefarmers)andusersofSRSsuchasICEandJASEC.

This boundary organization not only transfers information down to farmers, it also feedbacks

information ‘upwards’.Morespecifically, ASA‐Pacaya,dueto its localpresenceandsupportto

farmers’decisionson soilmanagement, feeds informationback to scientistsandpolicy‐makers

supporting 1) identification of new research needs and 2) data recompilation on impacts of

climateextremesonSRSneededtoreduceuncertaintiesinglobalchangeimpactstudies(Ayensu

etal.,1999).Finally,experienceintheUSprovesthatstabilityandresourceendowmentforlocal

technical assistance services are keys to support the monitoring of complex ecosystem

degradationproblems,andtoidentifykeyresearchareaswithlocalstakeholdersespeciallyunder

climatechange(Joyceetal.,2003).Theseauthorsstresstheneedtoinstitutionalizethesetypes

of services to improve theadaptive capacityof society.Nevertheless, institutionalizationmight

behinderedbyactualresourceendowment,aresultofpastpolicydecisions.Indeed,asoutlined

by Eakin and Lemos (2006), the capacity to provide adaptive responses to environmental

degradationproblemof LatinAmericanStates (e.g. toprovide stable support for theboundary

workofagriculturalextensionoffices)hasbeenreducedbypastdecisions.Inourcase,theactual

resourceendowmentofASA‐Pacaya results fromdecisions to reduce publicexpenditure taken

undertheimplementationofStructuralAdjustmentprogramsbackinthe1980s(LeclercandHall,

2007). A limitation of our approach was determined by the static perspective we have on

information‐sharingnetwork. Indeed,SRSprovisionandusehappens in localcontextsandover

time.Moreover,actorsinvolvedininformation‐sharingnetworksmodifytheirunderstandingand

interests,andshapetheirpreferencesonalternativeactionsovertime.Furtherresearchwould

consideranalyzingthehistoricalevolutionofideas,actionsandpolicyissuesraisedovertimeasa

consequenceofnetworkinterchangeofinformationinCostaRica.

37Thisprojectanalysedfuturepotentiallandusescenariosandtheireffectsonon‐siteandoff‐siteeffectsonSRSprovision.

131

5.8Conclusions

Weaddressedtwoimportantaspectsofgovernancesystemforecosystemservicesunderclimate

change,namely: institutionalmismatchesandtheroleofboundaryorganizations.Weanalysed

acrossscalesthroughanempiricalstudy,formalregulationsandmandatesthatframeactionsof

relevant actors to design responses to SRS degradation and inter‐organization information‐

sharing networks. We identified important regulatory mismatches affecting access to scarce

financial resources. We also identified, through policy network analysis, important boundary

organizationsthathavepotentialtofosteradaptiveresponsesingovernancenetworks.However,

pastdecisionshaveweakenedtheoperationalcapacityofthisboundaryorganization.Ourstudy

suggeststhatinstitutionaleffortstostrengthentheboundaryroleplayedbytheextensionoffice

could benefit continuous learning which is a prerequisite of adaptive governance. Further

researchmightshedlightoni)howtheinformationtimelinessandresolutionscalesbydifferent

actors provides sufficient for organizations’ decision‐making processes, and ii) how networks

evolveovertimeandunderdifferentpolicyconstraintsandopportunities.

5.9References

Adger,W.N., 2003. Social capital, collective action, and adaptation to climate change. Economic Geography79(4),387‐404.

Adger,W.N., Brown, K., Tompkins, E.L., 2005. Thepolitical economyof cross‐scale networks in resource co‐management.EcologyandSociety10(2).

Agrawala, S., Broad, K., Guston, D.H., 2001. Integrating climate forecasts and societal decision making:challenges toan emergentboundaryorganization.Science, Technology,&HumanValues 26(4),454‐477.

Aguilar,E.T.,PetersonC.,Ramırez,P.,Frutos,R.,Retana,J.A.,Solera,M.,Soley,J.,GonzalezGarcıa, I.,Araujo,R.M., Rosa Santos, A., Valle, V.E., Brunet, M., Aguilar, L., Alvarez, L., Bautista, M., Castañonp, C.,Herrera,L.,Ruano,E.,Sinay,J.J.,Sanchez,E.,HernandezOviedo,G.I.,Obed,F.,Salgado,J.E.,Vazquez,J.L., Baca,M.,Gutierrez,M.,Centella, C.,Espinosa, J.,Martınez,D.,Olmedo, B.,OjedaEspinoza,C.E.,Nuñez, R., Haylock, M., Benavides, H., Mayorga,R., 2005. Changes in precipitation and temperatureextremesinCentralAmericaandnorthernSouthAmerica,1961‐2003.JournalofGeophysicalResearch,Vol.110,D23107.

Ayensu, E., Van R. Claasen,D., Collins,M.,Dearing, A., Fresco, L., Gadgil,M., Gitay,H., Glaser, G., Juma, C.,Krebs, J., Lenton, R., Lubchenco, J., McNeely, J.A., Mooney, H.A., Pinstrup‐Anderson, P., Ramos, M.,Raven,P.,Reid,W.V.,Samper,C.,Sarukhas,J.,Schei,P.,Tundisi,J.G.,Watson,R.T.,Guanhua,X.,Zakri,A.H.,1999.Internationalecosystemassessment.Science286,685‐686.

Bodin,O.,Crona,B.I.,2009.Theroleofsocialnetworksinnaturalresourcegovernance:whyrelationalpatternsmakeadifference.GlobalenvironmentalChange19,366‐374.

Borgatti, S.P., Everett,M.G., Freeman, L.C., 2002.Ucinet forWindows: software for social network analysis,AnalyticTechnologies,Harvard,MA.

132

BorgattiS.2009.Networkanalysisinthesocialsciences.Science,323,892‐895.Brandes, U., Kenis, P., Raab, J., Schneider, V.,Wagner,D., Explorations into visualizationof policy networks.

JournalofTheoreticalPolitics11(1),75‐106.Brass,D.J.,Galaskiewicz,J.,Greve,H.R.,Tsai,W.,2004.Takingstockofnetworksandorganizations:amultivelel

perspective.AcademyofManagementJournal24(6),795‐817.Brinkerhoff,D.W., Coordination issues in policy implementationnetworks: an illustration fromMadagascar’s

EnvironmentalActionPlan,WorldDevelopment24(9),1497‐1510.Brooke, C., 2002. Climate change, vulnerability and conservation in Costa Rica: An investigation of impacts,

adaptive environmental management and national adaptation networks. PhD thesis, Oriel College,Oxford,UK,253p.

Brooke, C., 2002. Climate change, vulnerability and conservation in Costa Rica: An investigation of impacts,adaptive environmental management and national adaptation networks. PhD Thesis, School ofGeographyandtheEnvironment,OrielCollege,OxfordUniversity,Oxford,UK..

Burt,R.S.,2005.Brokarageandclosure:anintroductiontosocialcapital.OxfordUniversityPress,UK,279pp.Carlsson,L.,Sandstrom,A.,2008.Networkgovernanceofthecommons.InternationalJournaloftheCommons

2(1),33‐54.Cash, D.W., 2001. In order to aid in diffusing useful and practical information: agricultural extension and

boundaryorganizations”.Science,TechnologyandHumanValues26(4),431‐453.Cash,D.W.,MoserS.C.,2000.Linkingglobalandlocal scales:designingdynamicassessmentandmanagement

processes.GlobalEnvironmentalChange10,109‐120.Cash, D.W., Clark, W.C., Alcock, F., Dickson, N.M., Eckley, N., Guston, D.H., Jager, J., Mitchell, R.B., 2003.

Knowledge systems for sustainable development. Proceedings of the National Academy of Science(PNAS)100(14),8086‐8091.

Cowie, G.M., Borrett, S.R., 2005. Institutional perspectives on participation and information in watermanagement.EnvironmentalModelling&software20,469‐483.

Diaz, A., Locatelli,B., Vignola, R., Prins, C., 2009. Evolución de instituciones adaptativas frente a problemashidrológicos endos cuencas de Costa Rica. Submitted toRevista de RecursosNaturales y Ambiente,CATIE,CostaRica.

DiMaggio,P.J.,Powell,W.W.,1983.Theironcagerevisited:Institutionalisomorphismandcollectiverationalityinorganizationalfields.AmericanSociologicalReview48,147–160.

Duit,A.,Galaz,V.,2008.Governanceandcomplexity:emergingissuesforgovernancetheory.Governance:aninternationaljournalofPolicy,AdministrationandInstitutions21(3),311‐335.

Eakin,H., Lemos,M.C.,2006.Adaptationand thestate: LatinAmericaand thechallengeofcapacity‐buildingunderglobalization.GlobalEnvironmentalChange16,7‐18.

Eastaugh,C.,Reyer,C.,Gonzalez‐Moreno,P.,Jian,W.,Biscaia,A.G.,Pentelkina,O.,2009.Forestagencies’earlyadaptations to climate change. International Union of Forest Research and Organizations (IUFRO)OccasionalPaper23,ISSN1024‐414X.

Engel,S.,Pagiola,S.,Wunder,S.,2008.Designingpaymentsforenvironmentalservicesintheoryandpractice:Anoverviewoftheissues.EcologicalEconomics65(4),663‐674.

Farquharson,K.,2005.Adifferentkindofsnowball:identifyingkeypolicymakers.InternationalJournalofSocialResearchMethodology8(4),345‐353.

Fischlin, A., Midgley, G.F., Price, J.T., Leemans, R., Gopal, B., Turley, C., Rounsevell, M.D.A., Dube, O.P.,Tarazona, J., Velichko, A.A., 2007: Ecosystems, their properties, goods, and services. Climate Change2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the FourthAssessmentReportof the Intergovernmental PanelonClimateChange,M.L.Parry,O.F.Canziani, J.P.Palutikof,P.J.vanderLindenandC.E.Hanson,Eds.,CambridgeUniversityPress,Cambridge,211‐272.

Fisher,B.,Turner,R.K.,Morling,P.,Definingandclassifyingecosystemservicesfordecisionmaking.EcologicalEconomics68,643‐653.

133

Folke, C., Hahn, T., Olsson, P., Norberg, J., 2005. Adaptive governance of social‐ecological Systems. AnnualReveiwofEnvironmentalResources30,441‐473.

Freeman,L.C.,(1977).Asetofmeasuresofcentralitybasedonbetweenness.Sociometry40(1),35‐41.Guo, Z., Xiao, X., Li, D., 2000. An assessment of ecosystem services: water flow and hydroelectric power

production.Ecologicalapplications10,925‐936.Guston. D.H., 2001. Boundary organizations in environmental policy and science: an introduction. Science,

Technology,&HumanValues26(4),399‐408.Hodge, I., 2001. Beyond agri‐environmental policy: towards an alternative model of rural environmental

governance.LandUsePolicy18,99‐111.Howlett, M.H., Ramesh, M., 1998. Policy subsystem configurations and policy change: operationalizing the

postpositivistanalysisofthepoliticsofthepolicyprocess.PolicyStudiesJournal26(3),466‐481.Hanneman, R.A., Riddle, M.,2005.Introduction to social network methods. Riverside, CA: University of

California,Riverside(availableathttp://faculty.ucr.edu/~hanneman/Huitema, D., Mostert, E., Egas, W., Moellenkamp, S., Pahl‐Wostl, C., Yalcin, R., 2009. Adaptive water

governance:assessingtheinstitutionalprescriptionsofadaptive(co‐)managementfromagovernanceprespective and defining a research agenda. Ecology and Society 14(1), Online:(www.ecologyandsociety.org/vol14/iss1/art26/

Hulme,P.E.,2005.Adaptingtoclimatechange:istherescopeforecologicalmanagement inthefaceofglobalthreat?JournalofAppliedEcology42,784‐794.

Instituto Nacional De Estadística y Censo Centroamericano de Población (INEC), 2002. Estimaciones yproyeccionesdepoblación1970‐2050.SanJosé,CostaRica,112pp.

InstitutoCostaricensedeElectricidad(ICE),2007.Plandeexpansióndelageneracióneléctrica:periodo2008‐2021. San Jose, Costa Rica 131 pp. Available at:http://www.grupoice.com/esp/ele/planinf/plan_exp.htm)

Jost,F.G.,Jacob,K.,2004.TheclimatechangepolicynetworkinGermany.EuropeanEnvironment14,1‐15.Joyce,L.A.,2003.Improvingtheflowofscientific informationacrosstheinterfaceofforestscienceandpolicy.

ForestPolicyandEconomics,5,339‐347.Kjær,A.M.,2004.Governance.CambridgeUniversityPress,UK,256pp.Knoke,D.,1990.Politicalnetworks:thestructuralperspective.CambridgeUniversityPress,290pp.Koiman,J.,1999.Social‐politicalgovernance:overview,reflectionsanddesign.PublicManagement1(1),67‐92.Leclerc,G.,Hall,C.A.S.,2007.MakingWorldDevelopmentwork:scientificalternativestoneoclassicaleconomic

theory,UNMpress,645pp.Levitt,B.,March,J.G.1988.Organizationallearning.InW.R.Scott&J.Blake(Eds.),Annualreviewofsociology,

PaloAlto,CA.Leydesdorff, L., 2007. Betweenness centrality as an indicator of the interdisciplinarity of scientific journals

58(9),1303‐1319.Magrin,G.,GayGarcía,C.,CruzChoque,D.,Giménez,J.C.,Moreno,A.R.,Nagy,G.J.,Nobre,C.,Villamizar,A.,

2007. Latin America. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution ofWorkingGroupIItotheFourthAssessmentReportoftheIntergovernmentalPanelonClimateChange,Parry,M.L.,Canziani,O.F.,Palutikof,J.P.,vanderLinden,P.J.,Hanson,C.E.,Eds.,CambridgeUniversityPress,Cambridge,UK,581‐615.

Mahanty, S., 2002. Conservation and development interventions as networks: the case of the IndiaEcodevelopmentproject,Karnataka.WorldDevelopment30(8),1369‐1386.

McDaniels,T., Longstaff,H.,Dowlatabadi,H.,2006.Avalue‐based framework for riskmanagementdecisionsinvolvingmultiplescales:asalmonaquacultureexample.EnvironmentalScienceandPolicy9,423‐438.

Millennium Ecosystem Assessment (MEA), 2005. Ecosystems and human well‐being: hydrological ecosystemservicesreport.IslandPress,WashingtonD.C.,USA.

Miranda,M., Monge,C.,Otoya,M., 2006.Formulacióndepolíticaspara laadaptacióndelsistemahídricoalcambio climático. Proyecto Fomento de las Capacidades para la Etapa ‐ II de Adaptación al Cambio

134

Climático en Centroamérica, México y Cuba. PNUD‐MINAE‐IMN. Centro Internacional de PolíticaEconómicadelaUniversidadNacional(CIMPE‐UNA).SanJosé,CostaRica.84p.

Ostrom,E.,2007.Adiagnosticapproachforgoingbeyondpanaceas.ProceedingsoftheNationalAcademyofScience(PNAS)104(39),15181‐15187.

Pagiola,S.,2008.PaymentsforEnvironmentalservicesinCostaRica.EcologicalEconomics65(4),712‐724.Pahl‐Wostl,C.,Sendzimir,J.,Jeffrey,J.,Arts,J.,Berkamp,G.,Cross,K.,2007.Managingchangetowardadaptive

watermanagementthroughsociallearning.EcologyandSociety12(2),art.30.Pelling,M., High, C., 2005.Understanding adaptation:what can social capital offer assessments of adaptive

capacity?GlobalEnvironmentalChange15,308‐319.Penker,M.,Wytrzens,H.K.,2008.Evaluatinglandscapegovernance:atoolforlegal–ecologicalassessments.In

Hadorn, H.H., Hoffmann‐Riem, H., Biber‐Klemm, S., Grossenbacher‐Mansuy, H., Joye, D., Pohl, C.,Wiesmann,U.,Zemp,E.,2008.HandbookofTrandisciplinaryResearch,SpringerEds.448pp.

Piattoni,S.,Informalgovernanceinstructuralpolicy.PerspectivesonEuropeanPoliticsandSociety7(1),56‐74.Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., McNair, M., Crist, S., Shpritz, L., Fitton, L.,


Ponce,L.O.,2006.Los instrumentoseconomicosenlagestiondelagua.ElcasodeCostaRica.UnitedNationsEconomicCommissionforLatinAmerica(ECLAC),D.F.,Mexico,59pp.

Raab,J.andKenis,P.(2006).TakingStockofPolicyNetworks:DotheyMatter?InFischer,F.,Miller,G.J.,andSidney,M.S.,editors,HandbookofPublicPolicyAnalysis:Theory,Methods,andPolitics,number125inPublicAdministrationandPublicPolicy.Taylor&FrancisCRCPress.

Rotmans,J.,Rothman,D.S.,2003.Scaling in integratedAssessment.SwetsandZeitlingerpublishers,Lisse,theNetherlands,355pp.

Segura,O.,2004.AgendaAmbientaldelAguaenCostaRica.RevistaGeográficadeAméricaCentral40,39‐50.Southgate, D., Macke, R., 1989. The downstream benefits of soil conservation in ThirdWorld Hydroelectric

Watersheds.LandEconomics65(1),38‐48.Toimu,I.,1999.Dataismorethanknowledge:implicationsofthereversedknowledgehierarchyforknowledge

managementandorganizationalmemory.JournalofManagementInformationSystems16(3),103‐117.Tuinstra,W.,Hordijk,L.,Kroeze,C.,Movingboundariesintransboundaryairpollutionco‐productionofscience

and policy under the convention on long range transboundary air pollution. Global EnvironmentalChange16(4),349‐363.

Turton, A.R., Hattingh, H.J., Maree, G.A., Roux, D.J., Claassen, M., Strydom, W.F., 2007. Governance as atrialogue:government‐society‐scienceintransition.Springer‐Verlang,Germany,354.

UittoJ.L.,1997.Environmentalgovernanceandtheimpendingwatercrisis.GlobalEnvironmentalChange7(2),167‐173.

Vander Brugge, R., van Raak, R., 2007. Facing the adaptivemanagement challenge; insights from transitionmanagement.EcologyandSociety12(2),art.33.

VanNoordwijk,M., Poulsen, J.G., Ericksen, P.J., 2004.Quantifying off‐site effects of land‐use change: filters,flowsandfallacies.Agriculture,EcosystemsandEnvironment104,19‐34.

vanNoordwijk,M.,Leimona,B.,Emerton,L.,Tomich,T.P.,Velarde,S.J.,Kallesoe,M.,Sekher,M.,Swallow,B.,2007. Criteria and indicators for environmental service compensation and reward mechanisms:realistic,voluntary,conditionalandpro‐poor.WorldForestryCentre,CESScopingStudyIssuepaperno.2.ICRAFworkingpaperno.37,Nairobi,Kenia.

Wasserman,S.,Faust,K.,1994.Socialnetworkanalysis:methodsandapplications.CambridgeUniversityPress,Boston,USA,825pp.

Weible,C.M.,Sabatier,P.A.,2005.Comparingpolicynetworks:marineprotectedareasinCalifornia.ThePolicyStudiesJournal33(2),181‐201.

Weible, C.M., 2008. Expert‐based information and policy subsystems: a review and synthesis. The PolicyStudiesJournal36(4),615‐635.

136

6. Conclusions

6.1Researchquestionsandtheirrationales

Thissectiondescribesandsummarizestheframeworkandscopeofthequestionsaddressed in

thisthesis.Thethesisinvestigateshuman‐environmentsystemdimensionsofsoilerosionandits

controlinCostaRica.Theanalysisofthehumansystemrelevantforsoilerosioncontrolrequires

theconsiderationofthedecisionmakingandinteractionsofactorsoperatingatdifferentscales.

Thisismainlyduetohowtheirdecisionsinfluencethespecificdynamicsofsoilerosionthatstarts

inupstreamareaswheretopsoilparticlesaredetachedand,throughwaterrunofftransportation

processes, flows down to downstream areas. Societal responses have been based on

assumptionsonactors’decisionsand interactionsalongtheseenvironmentalprocesses.Society

hashistoricallyassumedan inverse relationshipbetween forest cover inwatershedandwater‐

runoff and this has drivenmost efforts to protectwater resources andwatersheds in general.

However, systematizing peer‐review studies worldwide reveals that these decisions might be

based on uncertain assumptions on the relationship among upstream forest‐cover effects on

water‐relatedecosystemservicessuchascontrolofrunoff(andthuslaminarerosionprocesses).

On the other side, in Costa Rica societal collective responses, characterizing the human

componentofthesystem,haveassumedthatactorsdirectlybenefitingfromon‐siteandoff‐site

SRSsuchasupstreamfarmersanddownstreamhydropowerfacilityrespondmainlytoeconomic

incentives structures. The motivations of these two types of actors to get involved in SRS

managementdependonhowthedegradationofSRSaffectstheiractivities.Morespecifically,for

upstream farmers soil erosion is a concern due to its effects on decreasing soil fertility and

increasingtheneedtorestorethesoilnutrientslost.Ontheotherside,hydropowerfacilitiesare

concernedbythe increasingcostsofmanagingsedimentscomingfromupstreamareastotheir

dams. However, thedecisionsoftheseactorsaredrivenbytheir cognitiveandsocioeconomic

characteristics and by institutional settings in which they are embedded. Indeed, incentive

distributionmechanismsaswellasgeneralinstitutionalsettingshaveimportantconsequenceson

the constraints and opportunities faced by these actors in promoting soil conservation. This

thesishasaddressedquestionsarisingfromthesesocietalassumptionsbyanalyzingthehuman

137

andtheenvironmentsystemsofsoilerosionanditscontrolinthespecificcasestudyoftheBirris

watershedofCostaRica.

Chapter 2 concentrates on the questions related to the environmental consequences of

upstream forest‐cover change on watershed hydrological responses including water runoff

control(animportantdriveroftopsoilerosionanddownstreamsiltation.Firsttheenvironmental

characteristics of the relationship among upstream land uses and downstream hydrological

effectsincludingwaterrunoff,baseflowsandtheregulationofwatercyclewasanalyzed.Then,

in an aggregated meta‐analysis evidences from worldwide watershed experiments on how

upstream complex land‐usemosaics influence hydrological responses influencingwater runoff

andconsequentlytop‐soilerosionwereanalyzed.

Inthefollowingchapters,questionsofthethesisfocusontheconstraintsandopportunitiesfor

the human system to organize collective responses to promote upstream soil conservation

measures (i.e. thus reducing water runoff and its consequences on downstream users of soil

regulation services).Researchquestionsand focus follows themultiple scalesatwhich specific

decision‐makingprocessesarerequired.

So,thequestionsofrelevanceforChapter3focusondecision‐makingoflocalupstreamfarmers.

Theirdecisionsonsoilmanagementdirectly interactwithaclimatic (i.e.precipitation intensity)

and site‐specific (i.e. topography and soil types) to determine the level of provision of soil

regulationservicestodownstreamarea.Thespecificquestionsrefertohowterritorial,cognitive

and socioeconomic aspects combine to characterize different voluntary efforts by farmers to

controlsoilerosion.Thisaddressesthehypothesisthatmainlysocioeconomicaspectsdetermine

the level of conservation efforts implemented by farmerswhile cognitive aspects (such as risk

perceptionsoferosion,valuesandknowledgeoncauses,effectsandsolutions)donot.

Moving at scale of the whole watershed, the analysis of decision‐making processes need

including the analysis of decision‐making and interaction (along the specific processes of

upstream topsoil loss and downstream transportation of sediments) of those directly involved

with the on‐site and off‐site effects of SRS degradation. In Chapter 4, the question on what

institutional characteristics should a collective response among these actors have is analyzed.

Although multiple objectives characterize the definition of common responses among these

138

actors, society has been driven by assumptions on the role of direct economic incentives (e.g.

Payment for Ecosystem Services) to promote socially desired mechanisms. The question

addressedinthischapterreferstowhatotherobjectives,inadditiontodirectpayment‐transfer,

compose a socially desired mechanism and how negotiation analysis can help structuring

collectiveresponsesaccountingforthoseobjectives.

Decisionsofactorsdirectly involvedwithon‐siteandoff‐site consequencesof SRSdegradation

areconstrainedandfosteredbytheiraccesstoproperregulationframeworkandinformation.In

thisrespect,decision‐makingofdifferentactors(i.e.otherthanthosedirectlyinvolvedwithSRS

provision)hasdirectorindirectinfluenceoncollectiveactionstoreduceSRSdegradation.Indeed,

this direct and indirect influence happens in a complexity of contrasting objectives, different

knowledge systems and different information‐needs and ‐exchanges among actors. In this

complex multi‐actors’ setting, research questions for Chapter 5 refer to i) which institutional

mismatchesandgapsconditiondecisionsofthosedirectlyinvolvedinSRSprovisionandbenefits

and ii) which actors play a an important boundary role across scales and knowledge systems

providingspacetoaddressthosemismatchesandgaps.

Finally,intheAnnex2,Idonotaddressaspecificresearchquestionbutpresenttheresultsofa

consultation involvingmore than 80 stakeholders from LatinAmerica involved in research and

policies on ecosystem services and adaptation to climate change. This consultation shows an

interesting match among findings of the thesis and stakeholders’ concerns on the role of

differentactors (i.e.directlyor indirectly involvedwithprovisionanduseofecosystemservices

underclimatechange).

6.2Unansweredresearchquestions,limitationsandfutureresearch

The high complexity of the human‐environment system characterizing SRS on‐ and off‐site

provisionchallengestrandisciplinaryresearchandleavesspacetomanyun‐answeredquestions.

Weherehighlightthemostimportantaspectsthatcouldnotbeaddressedinthisthesisandthat

can be object of future research efforts. Starting with the analysis of the environment

component, we could not include, in our meta‐analysis, paired‐catchment experiments

considering comparisonof runoff responsesofagriculturalwatershedswithdifferenteffortsof

soilmanagementinproductionunits.Increasingevidencesfromaroundtheworldinthisrespect

139

couldimprovealsoinformedtargetingofsoilconservationpoliciesinareasofprioritywatersheds

in developing nations.Moreover, ourmeta‐analysis could not benefit of comparison of forest

covereffect inwatershedswithdifferent levelofsoildegradationasawaytodemonstratethe

beneficialroleofforestinotherregulationservicesassociated,forexample,towaterinfiltration.

The following three papers on the human component, we also outline some un‐answered

research questions on the factors that hinder or promote how individuals’ or collective’s

responsestoSRSdegradationaredesignedandimplemented.

Attheindividualfarmerlevel,ouranalysishasidentified,basedoncombinationofcognitiveand

socioeconomicvariables,groupingsoffarmersaccordingtotheirconservationefforts.Themodel

used expanded the traditional econometric analysis of soil conservation adoption decisions by

addingriskperceptionandknowledge.However,forfutureresearchanalysisoffarmersinternal

networkforinformation‐sharingmightrevealinterestingprocessesofinformationdissemination

thatcomplementandupdatetheirknowledgesoastheirperceptionoftheriskfactorsaffecting

SRSdegradation.

At thewatershed scale level the two‐parties negotiation analysis is only partially covering the

issuesthatmustbetakenintoaccountSRSmanagement.Indeed,spatialaspectsareparticularly

relevantindefiningstrategiestopromotecollectiveactionsespeciallyregardingtheprovisionof

off‐site benefits to downstream actors. Here, more research is required on what strategy is

sociallyacceptableinagivencontexttopromotecollaborationoffarmersbyerosionriskpriority

areas.Morespecifically,soilmanagementdecisionsoffarmersinupperstreamareasdetermine

theefficacyofsoilmanagementpracticesinlowerstreamzones.Inthisrespect,futureresearch

might be interested in looking at the most‐preferred incentives mechanisms to promote

negotiations/cooperationamong farmersaccording to their relative spatial importanceof their

soilmanagementdecisions.Intheanalysisofcross‐scaleinformationnetworkinteractionsfuture

research in the benefit of adaptation planning could analyse the stability of networks under

different sectorial policies that affect objectives and interests of actors to interact and share

information.Consideringthattheroleofextensionofficesisstrategicinadaptivegovernanceof

SRS,athoroughanalysisofincurredcostsandexpectedbenefitsofinvestingintheseservicesin

the face of climate change projected increase in erosive extreme precipitation events might

140

promoterevisionofpastdecisionsthatreducedtheiravailableresourcesandcouldbepartofan

analysisofthebenefitsofSRSprotectioninadaptationplanninginCostaRica.

6.3Mainfindings

In the environment system analysis we could show that, systematizing different studies

comparing theeffectofnatural forest coverandplantations (NFCP)vs. non‐forest cover (Non‐

FC), there is no significant difference in the contribution of these two groups of land uses in

controlling storm flow (Sections 2.3.2 and 2.3.3). Since this hydrological response is a strong

proxyofwaterrunoffandthussoilerosionresultsindicatethatimprovedsoil‐managementalso

inlandswithlowornoforest‐cover(suchasagriculturalandlivestockproductiveunitswithsoil

andwaterconservationpractices)canalsocontributetocontrollingsoilerosion.

Inupstreamlocations,traditionalsocioeconomicvariablesareonlypartiallyinfluencingfarmers’

decisionmakingonsoilconservation.Consistentlywithpreviousliterature,ourfindings(Section

3.4.4) farmsize canbean importantdeterminantof soil conservationbehaviour.Results show

that this is especially true in the case of small farmers where, although showing higher

perception of erosion risk, farm land has a high opportunity costs and the benefits of soil

conservation practices might be perceived too little. In addition to traditional socioeconomic

variables,ourdecisionmodel indicatesthat farmers’awarenessofsoilconservationcausesand

impactsisalsoaprerequisiteofsoilconservationbehaviouralthoughnotsufficient.Indeed,the

locationofproducersinhigherosionareasandinareascoveredbysoilconservationprogramisa

strongdeterminantofsoilconservationbehaviour.Similarly,higherriskperceptionsontherole

ofhumanactivitiesandnegativevalueson theextent towhich technologycanhelpovercome

consequences of topsoil loss are importantly associated to higher farmers’ soil conservation

efforts in upstream areas. Thus, expanding access to programs raising awareness on the

consequencesofhumanactivitiesandlimitationsoftechnologiesinaddressingtheeffectsofsoil

erosioncanproveahighlycost‐effectivestrategytoincreaseSRSprovisionbothon‐siteandoff‐

site.

Atthewatershedscale,applyingnegotiationanalysismethodshelpedoutliningandstructuring

the multiple objectives characterizing parties directly involved with SRS on‐site and off‐site

provision. We could synthesize and present formally and jointly preferences of SRS directly‐

141

relatedactorsover thecharacteristicsofa soil conservationmechanism.Our results show that

key governance aspects such as types of incentives, fairness in distribution of incentives, and

trust in intermediary are essential component to design socially‐desiredmechanisms (Sections

4.5.1.1 and 4.5.1.2). Direct payments for ecosystem services (highly promoted in national

watershedconservationprograms)appearasoneoftheleastpreferred incentivetype.Instead,

technical assistance is seen as an appropriate way to maintain productive activities while

protecting SRS. Intermediary actors are needed to mediate (i.e. to reduce transaction costs)

amongtheincentive‐giversandtheirintendedbeneficiariesthataresupposedtoimplementsoil

conservation.

The analysis of formal policies and mandates (Section 5.6.1) that across scales constrain or

promoteactors’ responses toSRSdegradationhighlighted that institutionalmismatcheshinder

the possibility of farmers and hydropower to access additional resources to promote their

collaborativeefforts.Inthisrespect,formallawsrequireusersofoff‐siteSRSsuchashydropower

companies to demonstrate the real costs and benefits incurred by activities to protect their

priority watersheds as a prerequisite for including an environmental tariff in the electric bill.

However,datascarcityanddeepuncertaintycharacterizesuchstudiesespecially inthecontext

of climate change impacts on soils in the future and possible responses.We usedmulti‐actor

analysis to analyze parameters of information‐sharing network which help highlighting

organizations that play a key role in disseminating information and linking actors at different

scaleandpartofdifferentknowledgecommunitiesinvolveddirectlyorindirectlyinSRSprovision

(Section5.6.2).Concordantlywithprevious research thisanalysis finds that the localextension

office is connecting actors interested in on‐site and off‐site benefits of SRS.Moreover, across

scalestheylinkalsotothescientificcommunitiesthatdesignandimplementresearchtorespond

toSRSdegradation.Thisrole isstrategic inadaptivegovernanceofSRSandnationalregulators

shouldreconsidertheirpastdecisionsinreducingthebudgetexpenseinfavoroftheseservices.

6.4LessonslearnedforothercasesonESdegradation

In this section, the thesis presents conclusions drawn from the Birris watershed case

study on SRS that can be applied elsewhere in ES or specifically SRS contexts of

developing countries. The Birris case study is defined by its strongly market‐oriented

142

agriculture combined with a high population density dependent on land use‐based

activities. Moreover, this context can also be characterized, along with the IPCC

definition,byitsvulnerablecontext.Wecanidentifyprocessesthatincreaseexposureas

indicatedbyobservedandprojectedincreaseinthenumberand/orintensityofextreme

precipitation events.Moreover, agricultural activities pose strong pressure on few‐left

forestresourcesand,duealsotosteepslopes,soilresourcesincreasingsensitivityofthe

systemtoerosion.Basedontheselargelydefineddimensionsofourcasestudywecan

draw lessons that might be relevant for other contexts where society in a given

landscapestronglydependsontheprovisionofecosystemservicesthatarethreatened

bycombined societal and climatechange impacts.These lessons refer to themethods

applied in the thesis to describe, analyze and prescribe actions for planning collective

actionstorespondtoESdegradation.

In developing world, water‐related ecosystem services are at the centre of different

initiatives aiming at sustaining development (e.g. Millennium Development Goals;

national hydroelectricity policies). Protecting or restoring forest cover for its role in

water‐relatedecosystemservices inspecificwatersheds isconsideredanessentialpart

ofthesecollectiveeffortsalthoughtheneededscientificandcontext‐specificevidenceto

makeinformeddecisionisoftenmissing.Generally,thistypeofinformation iscostlyto

produce, requires large series of data and, especially considering the complex climate

change alteration of hydrological cycle, require thorough analysis and clear

communication of uncertainties. In the light of these considerations, some important

hintscanbegivenforwatershedsofdevelopingcountrieswhere,astheBirriscasestudy,

largedeforestationhasalreadyoccurred,market‐orientedagricultureiswellestablished

andhighfragmentationofagriculturalproductiveplotscanbefound. Inthesetypesof

watersheds, it is difficult to hypothesize the design and implementation of large and

extended researchproject to study thehydrologicalbenefitsof increasing forest cover

(needed to provide evidence for improved water‐related services). Limitations are

relatedtothepresenceof importantagriculturalactivitiesthatguarantee livelihoodof

local inhabitants. Thus, considering the above contextual limitations and the

uncertaintiesontheeffectivehydrologicalbenefitsof increasingtreecover,watershed

143

managementplanshouldstrivetofosterindividualandcollectiveresponsestopromote

the role of agriculture in the provision of on‐ and off‐site ecosystem services. Indeed,

agricultural production system can potentially reduce erosion and dam’s siltation by

introducingtechniqueslargelyknowntoagriculturalscientistsandtomanyfarmers.

In this context, the provision of ecosystem services by agricultural systems can be

promoted through a combination of complementary policies. While direct economic

incentives such as PES schemes has been widely promoted as a means to stimulate

individualconservationbehaviour, ithasbeensubjectofseveralcriticsrelatedtosocial

ethics, environmental performance, transaction costs and uncertain opportunity costs

miningitseconomicefficiency.LessonslearnedfromtheBirriscasestudyandthereview

of results from other cases show that policies and programs supporting technical

assistance aswell as awareness‐raising on causes and consequences of environmental

degradation in agricultural landscapes can prove a strong determinant for voluntary

conservationbehaviour.

The methods and tools used to analyze the cross‐scale decision processes of SRS

managementisapplicabletootherSRSwatershedcontextsindevelopingworldaswell

astootherecosystemservicesmanagementsituation.Inthisrespect,individualfarmers’

decision‐making can be studied through a combination of social, economic, territorial

and cognitive aspects. To be broadly applicable to different ES contexts, these

dimensions need to be adapted to reflect the specific barriers and opportunities that

farmersmightfaceinthefaceand/orperceiveofanESotherthanSRS.Forexample,for

the case of global ecosystem services such as climate regulation, the perceived

opportunitiesand institutionalbarrierstoaccess incentives(i.e.transactioncosts)from

carbonmarketsmightbeaveryimportantcognitivefactorinfluencingfarmers’decision

toincreasetreecoverintheirfarms.Movingatlandscapescale,multipleactorsinfluence

directlythelandusedecisionsaffectingtheprovisionofESinaspecificarea.Themethod

proposedinthenegotiationanalysischapterofthethesisopenstheopportunityforjoint

gainsthroughtheanalysisoftheseactors’objectivesandcreationofalternativesshaped

byconverginginterestsandvalues.Apossiblechallengeintheapplicationofthismethod

144

inotherEScontextsmightresideonthesynthesisofi)theobjectivesofmultipleactors

intocreativealternativestoproposefornegotiationofmechanismcharacteristics;andii)

the resultof thepreferenceelicitationexercise.This isespecially thecaseofglobal ES

whereoftenlandowners(especiallyforestcommunities)providingtheserviceofforest

carbonsinks liveawayfrompotentialbuyersandthesetwoactorshavenormally little

experienceofinteractionamongthem.

6.5 Keymessages

Responses to the degradation of Soil Regulation Services require consideration of decision

processeshappeningatmultiplescalesandfacinglargeuncertaintiesonthecombinedeffectsof

landmanagementandclimatechange.Thefollowingthesis’keymessagessuggestactionsthat,

atdifferentscales,canpromoteSRSconservation:

1) Attheindividualfarmerlevel:

a. Promoteawarenessonimpactsproducedbytheinteractionoflandmanagementand

climatechangeamongactorsaffectedbySRSdegradationon‐andoff‐site.

b. Promote support to farmers’ soil conservation behaviour tailored to their specific

productioncontexts(e.g.farmsize).

2) Atthewatershedscale,soilconservationprogramsshouldusenegotiationanalysistoidentify

key componentsofa soil conservationprogramaccording tobothon‐and off‐siteusersof

SRS.Keycomponentsofadaptationresponsesshouldincludepromotionoflearningovertime

(throughtechnicalassistanceandmonitoring)andcost‐effectivedistributionofsupport (i.e.

focusingonerosionpriorityareas).

3) Finally, national agricultural policies in vulnerable regions should consider strengthening

resourcesendowed toagriculturalextension servicesas thesecan representakeyactor in

the implementation of SRS adaptive management across‐scales (i.e. bridging information

fromfarmerstonationalscientificandpolicy‐makingcommunity).

146

7.Annexes

147

Annex 1: Extended acronyms of organizations in the network and their specific mandates

PolicyArea Scale Organization

MandaterelevantforBirriscasestudy

Source

Regulatory National Authority

RegulatingPublicServices(ARESEP)

Ensureadequatequalityofpublicservices

includingwaterprovisiontodomestic,agriculturalandhydroelectricusers.

LawoftheRegulatoryAuthorityforPublicServices.Articles:5,6a)yb),9,23y25

Regulatetariffsofpublicservicessuchaselectricbillanddefineitemscoveredbytariffson

electricserviceincludinginternalizationofenvironmental

externalitiesinwatershedmanagement.

RegulationoftheARESEPLaw,Art.5,6a),9,34,47‐52

Legitimizetheinclusionofcostsofwatershedmanagementforhydroelectricity

productioninelectrictariff.

LawauthorizingElectricProduction,Art.5,6,24

Regulate,fiscalandfinancialauditingofpublicserviceproviderssuchas

hydropower.

Nationaltolocal

Waterdepartment‐

Ministryof

Environment

(MINAE)

ØGeneralLawofWater

AdministerallwaterresourcesintheCountry.

ØFrameworkLawfortheEnvironment,Art.50

Leadprocessesfordefiningnationalpolicies

onwater.

ØCodeforMines,Art.4.

Enforcemonitoring,controland

administrationofwaterinthewatershed.

ØRegulatoryLawforPublicServices.TransitorioIV.

Transactnationalconcessionsfor

hydropowerproduction.

Decree26635‐MINAE.TransfertheWaterDept.TotheNationalMeteorologicalInstitute

Operateasinformationdeskforindividualsand

148

Operateasinformationdeskforindividualsand

governmentorganizations.

Auditwaterandprovidesconcessionstousers.

ChairtheAdvisoryBoardonWaterandcoordinatememberinstitutions.

Chargewatertaxtousers(includinghydropower)anddefinemechanismsforidentificationof

priorityareastoinvestforESprovision.

Nationaltolocal

Ministryof

Agricultureand

Livestock(MAG)

Promoteresearchonhydrological,

hydrogeological,agriculturalactivitiesandtheirrelationwithsurfaceandundergroundwater.

1949CreationoftheMAG

Certificaterationalsoilandwateruseinirrigation

concessions.

ØLawN°7779forUse,ManagementandConservationofSoil,Art.6y33

Auditthroughtechnicalstudiestheappropriatemanagementofsoiland

waterandtransmitinformationtoMinistryofEnvironment(MINAE)forcorrespondenteventual

cancellationofconcessions.

ØRegulatoryLawforUse,ManagementandConservationofSoil

Audit,evaluatetechnicalstudiesfordefinitionofappropriateagriculturalusesandzoningofrural

areastodefineappropriatemanagement

accordingtoenvironmental,socialand

agriculturaluse.

ØRegulationfortheControlandRegistrationofDangerousProductsArt.4y12

Defineandcoordinateimplementationof

nationalsoilconservationplansincollaboration

withrelevantagriculturalorganizations.

RegulationforUse,RegistrationandControlofPesticidesArt.2y66

Implementresearchtoimprovelandand

resourcemanagement.

1995DecentralizationprocessunderWorldBankfinancingtoprivatizesomeservices

provisionandcreatingthreemaindivisions:Technologyresearch;ExtensionservicesindecentralizedBaseAgricultureCentres;and

Administration.

149

Promotetrainingtoalllevelsforincreasing

transferofappropriatetechnologiesforlanduseandmanagement;Thiswasformedforanalysis

anddiscussionofagriculturalpolicy‐making.

ThecreationoftheNationalInter‐organizationalForumforAgriculture,Executive

DecreeNº31170‐MAG,2003.

Providetechnicalassistancetoagriculturalproducerstopromote

appropriatelandmanagement.

ConsolidatedNationalNetworkforConservationistAgriculturewithorganizationsasMAG,ICE,ITCR,CNFL,UNED,UCR,UNA,tostrengthenresearchanddevelopmentof

conservationistagriculture. Providecriteriato

evaluateconsequencesofdifferentwateruses

includinghydropowerandagricultureonsoil

resources.

Keepregistersofandmonitororganizationsandindividualsimplementing

soilconservationactivities.

ElaboratedtheAgriculturalStrategy21withSupportofIICAofficeinCostaRica

Maintainadatabankwithregistersofsoil

managementandlandusecapacityincluding

informationonsocioeconomic,technical

andenvironmentalaspects.

TheNationalProgramtoPromoteSustainableAgriculturewasstartedwithIADBloan

1436/OC‐CR(MAG/BID)PFPAS.

Promoteconstantlyparticipationofcivil

societyinsoilconservationactivities

throughitsdecentralizedextensionofficessuchas

ASA‐MAGPacaya.

National NationalFundforForestry(FONAFIFO)

DecentralizedinstitutioninsidetheNational

ForestryAdministrationwithmandateto:

Law71741990(ExecutiveDecreeN.19886‐MIRENEM;

Fundforwatershedandecosystemconservationactivitiestosmallandmediumlandusers;

NormN032,Law7216,1991,Creationofthefund

Fundreforestation,agro‐forestryandregenerationofnaturalvegetationin

priorityareasforwatershedprotection,biodiversity,carbonand

Law7575,Art.46creationofFONAFIFO

150

Fundreforestation,agro‐forestryandregenerationofnaturalvegetationin

priorityareasforwatershedprotection,biodiversity,carbonand

scenicbeauty;

Law7575,Art.46creationofFONAFIFO

Definecriteriaforidentificationofpriorityareaswherefundingshouldbetargeted;

Ensureappropriatefundingtopromote

provisionofappropriatelevelofES;

Defineandimplementmechanismsforfund

assignment.

Nationaltolocal

NationalSystemfor

ProtectedAreas(SINAC‐MINAE)

Decentralizedandparticipatoryorganization

integratingforestry,wildlifeandprotectedareasdepartmentsof

MINAE.

ØLawofCreationofNationalServiceforProtectedAreas,Art.13.

Establishallianceswithrelevantactorstopromotewatershed

protection.

ØBiodiversityLaw,Art.22,25y59.

Defineandenforcerulestoregulatelandusein

protectedareas.

FrameworkLawoftheEnvironment,Art.34.

Local Municipality

Takepartinthelocaladvisoryboard.

ØDecreeN.26635‐MINAE:Art..5y7CreationandFunctionoftheAdvisoryEntityforWater

Promoteprocessesoforganizations

implementingsoilconservationactivities.

ØWaterLaw,Art.31,41,154‐158,177,194‐198

Enforcelocaldecreesandconcessionstoprotect

forestsandwatersourcesinwatershedplanning.

ØDecreeN°26624‐MINAE

Evaluateandprovideapprovalofwatershedmanagementplans.

ØMunicipalCode,Art.4c),6,13d),79y81

Design,planandimplementpoliciesforsustainablenatural

resourcemanagement.

ØGeneralDrinkingWaterLaw,Art.4,5,6,710y11.

Recommendappropriatelegalenforcementfor

protectedareas.

ØLawofCreationofAYAArt.20y22

Enforcerulescorrespondingto

protectionstatusofnaturalareasthrough

decentralizedofficessuchasACCVC‐Cartagointhe

BirrisWatershed,.

ØFrameworkLawoftheEnvironment,Art.60

151

Protectandconservewatershedsandwater

resources.

ØLawofUrbanPlanning,Art.15y28.Establishmunicipalenforcementtoregulateurban

planninganddevelopment.Thisisrelatedtotheart.52oftheBiodiversityLawand28ofthe

FrameworkLawfortheEnvironment LawforConstructionArt.1and3providing

guidanceonconstructionofinfrastructures ØDecreeN.26635‐MINAE:Art..5y7Creation

andFunctionoftheAdvisoryEntityforWater ØWaterLaw,Art.31,41,154‐158,177,194‐

198 ØDecreeN°26624‐MINAE ØMunicipalCode,Art.4c),6,13d),79y81 ØGeneralDrinkingWaterLaw,Art.4,5,6,7

10y11. ØLawcreatingAYA,Art.20y22 ØFrameworkLawoftheEnvironmentArt.60 LeydeConstruccionesArt.1y3:Sobrelos

permisosparalaconstrucción National National

Registerof

Agronomy

Engineers(CIA)

Audittheprofessionalworkofagronomic

engineersinCostaRicatoensurethatadequateuseofnationalresourcesis

promoted.

Law7121,1991;creationofCIA

Regulateagronomicengineers’operationto

ensuretheyaredeliveringappropriatetechnical

assistancetoagriculturalproducers.

ModificationofArt.52and59ofLaw7121

National NationalServicefor

undergroundwaters,irrigation

anddrainageSENARA(MAG)

AutonomousorganizationaffiliatedtoMAGandwithmandateto

implementtechnicalstudiesonsurfaceandundergroundwaterstomonitoradequateuse.Providestechnical

informationtoARESEPforestimationoftariffsbased

onconsumption.

ØLawN°6877CreationoftheNationalServiceforUndergroundWater,Art.3

Provideadvisoryservicestoindividualsandpublic

organizations.

ØGeneralRegulationforSENARA.

ProvidetechnicaladvisoryservicestoMINAEto

locatewatersourcestobeprotected.

ØRegulationforSENARAcontrolofirrigationwáter

Supervisepolicyprocessestoformulateconcessionsonirrigation

waterstoensure

BiodiversityLaw,Art.114

152

transparency. Promotesoilandwater

conservationinirrigatedareas

Implementconstructionofsoilandwaterconservationinfrastructures.

Demand Nationaltowatershed

NationalInstitute

forElectricity(ICE)and

departments

suchasPISA,

Environment,Projectsand

decentralizedofficessuchasthe

Reventazon

Watershed

ManagementUnit

(UMCRE)

Developnationalresourcesforenergy

productionwithemphasisonhydropower.

ØLawN°449forthecreationoftheNationalInstituteforElectricity(ICE),Art.1,2,16

demandingtofosterexploitationofenergysourcesinthecountry,whilesustainablyusing

hydropowerpotentialandprotectingwatersheds.

Provideforrationaluseofwaterresourcesforenergyproduction.

LawN°7200authorizingprivategenerationofelectricity,Art.7

Conservewaterresourcesthroughactivitiesforwatershedprotection.

UMCREcreatedaftertheWatershedManagementPlanfundedbyIADB,2000.

Watershed CartagoAdministrative

Boardonwater(JASEC)

Hydropoweruserofecosystemwaterservices,benefitsfromupstream

erosioncontrolforoperatingitsdownstream

plants.

Law3300,1964;creationoftheBoardtoadministrateMunicipalElectricCompany

Promotesustainableuseofwaterresourcesinthe

Birriswatershed.

Law7420,1994;ConcessionforusingwaterforhydropowergenerationinBirrisWatershed

Trainingtoschoolsandproducersonbenefitsof

reforestation.

Law8345,2003;AllowingJASECtooperatehydropowerplantsallovernationalterritory

FormalAgreementwithICE‐UMCREforfarmers'

153

schoolsinpriorityareas,2003. PromoteEstablishthe

watershedmanagementplanthroughformationofacommitteetofosterwatershedprotectionthroughsoilandwaterconservationCollaborate

withotherlocalorganizationssuchasASA‐MAGPacayatoimplementtheplan.

Supply Watershed CommissionforWatersh

edplanning

inUpstrea

mReventaz

onWatersh

ed(COMCU

RE)

Elaborate,implementandcontrolthewatershedmanagementplanforUpstreamReventazon

Watershedwithemphasisonwaterservices.

LawN°8023PlanningandManagementoftheUpperReventazonWatershed,Art.4,5,6

Defineandimplementtrainingactivitieswithcommunitiesandpersonnelof

organizationsonwatershedplanning.

Implementactivitiesinpriorityareasprovidingecosystemservices.

Local Producers

associations

(ADAPEX,ADICO,COOPEBAIRES)

Personalinterviews

Promoteactivitiestoenhancemarketandfinancialaccessof

producers.

Promotetrainingofproducersoninnovativetechniquestosustainablymanagesoilinfarms.

National NationalInstitute

for

Autonomouspublicinstitutionwithmandatetoimplementnational

Law6735,1982;CreationofIDA

154

Agriculture

Development(IDA)

agriculturalpoliciesapprovedbyNational

DevelopmentPlanofthenationalEconomicPolicy

PlanningBureau. Providelandandtenure

securitytoproducersLaw2825,1982;RegulationofLanduseto

fosterESprovisionandtenurerights Throughitsdecentralized

officespromotetechnologicalinnovationthroughtrainingonsoilmanagementtechniques,

economicandorganizationof

agriculturalproductionunits.

Nationaltowatershed

NationalCouncilfor

agricultural

production(CNP)

AutonomousinstituteascribedtoMAG,with

mandateto:

LawoftheNationalProducersCouncilreformedbyLaw6050andbyLawnº7742.

Fosteractivitiesthatincreasemarketandfinancialaccessof

producers;

Promoteactivitiesandstrategiesthatimprove

nationalfoodsecurityandsocioeconomicwell‐being

ofproducers;

Supportdisseminationoftechnologicalinnovations

inproduction,industrializationand

marketing;

Supportimportandexportofagriculturalproducersfocussingon

smalltomediumfarmers;

Establishpartnershipswithextensionandtechnicalassistance

programs;

National NationalInstitute

forSocialWelfare(IMAS‐regional)

LawNº4760CreatingtheMixedInstituteforSocialSupportanditsregulationdecreeNº

26940‐MIVAH‐MTSS

Formulateandimplementspolicy

Law8563tosupporttheMixedInstituteforSocialSupporttostrengthenIMASbyuseof

155

guidelinestocombatpovertyincluding

financinginitiativesthatthroughappropriatelandmanagementpromotesocialwelfareand

environmentalbenefits.

tax‐sourcesforpovertyreduction

Supportproducersaffectedbyclimate

extremes.

LawNº7742createstheprogramforagriculturalsectorproductionreformmodifiedbyLaw8563.Regulatesprovisionofresourcestosupportthenationalplanforagricultural

productionreconversion Supportwithsmall

subsidiespoorproducersthatincollaborationwithlocalASA‐MAGPacayaextensionofficeuseappropriatesoil

managementtechniques.

Establishallianceswithotherrelevantactorstoachieveitsgoalsofpovertyreduction.

Science Internationalto

National

RegionalCommitteeforwater

resources(CRRH)

Inter‐governmentalCommissionformedbyrepresentantofnationaltechnicalinstitutesonclimatechangeandhydrometeorologyof

CentralAmericanregion,promoting:sustainableuseandconservationofwaterresourcesthroughdisseminationoftechnical

studies.

Formalcreation1966

Communicatingtolargepublicandorganizationsonclimatechangeandvulnerabilitywater

resourcesanditsusers.

FormalAgreementwithRegionalIUCNheadquartertofosterinitiativesonclimatechangeadaptationinwaterresourcesand

supportregionalparticipationinCopenhagenCOP

Identify,coordinateandfacilitateprocessesand

programsrelatedtowaterresourcesinCentral

America.

Strengthenthelinksbetweenregionalprogramsand

internationalinitiativesdealingwithclimatechange,variabilityand

water.

Internationalto

RegionalInstitute

PartofOrganizationofAmericanStates,itisan

Creationin1948undertheOrganizationoftheAmericanStates(OAS)

156

Internationalto

National

RegionalInstitute

forAgricultu

reCooperation(IICA)

PartofOrganizationofAmericanStates,itisan

intergovernmentalorganizationwith

mandateto:

Creationin1948undertheOrganizationoftheAmericanStates(OAS)

Promotewell‐beingofruralpopulation;

In2000OASdecisiontomakeIICAthegoverningbodyoftheInter‐AmericanBoardforAgriculturewithmandatetoimproverurallifewithsustainableproductioninagriculture

Promotedialogueandconsensusonkeyissue

foragriculturaldevelopmentofLatin

America;

Disseminateresultsfromresearchcentresonagricultureandits

vulnerabilitytoclimatechangeimpacts;

Promotefundingforresearchorganizationstoimplementappropriatetechnicalresponsestoimpactsofclimate

extremes.

Internationalto

national

Regional/national

civilsociety(IUCN)

Promotenationalcampaignsonclimate

changeandadaptationinwaterresourcesmanagementwith

emphasisontheroleofecosystemsforsociety.

1948creationofIUCNasaninternationalNGOforNatureConservation

Disseminateinformationonclimatechangeandvariabilityimpactsonecosystemsandtheservicestheyprovide.

Supportandpromoteresearchtoimprove

knowledgeonecosystemandtheirservices.

Establishcollaborativepartnershipstofosterregulatorychangesforecosystemprotection.

Establishallianceswith

nationalandinternationalpartnerstofoster

research,transferanddisseminationofresultsforsustainableuseof

lands.

157

Nationaltolocal

NationalCommissionon

Emergencyandpreventionofrisks(CNE)

Inter‐organizationalcommissionformedbynineministersandthepresidentwithmandate

toorganizeandcoordinateemergency

responses.

LawN°8488providesCNEtheenforcementtopromotedisasterpreventionandpreparedness

Disseminateinformationonareasvulnerableto

climateextremesthroughdecentralizedlocal

committees.

Supportorganizationsfosteringwatershed

protectionactivitiesthatpreventnaturaldisasters.

National NationalInitiativeonPeacewith

Nature(IPN)

GovernmentalinitiativelideredbytheExecutiveOfficeofthePresidency

MinistrywithanPresidentialcommissioncomposedby25membersfromacademic,private,civilsocietyandpublic

organizations.

6June2008PresidentialLaunchoftheInitiativeIPNtocombatenvironmentaldegradation

Promotepolicyprocessestocounterarrestenvironmental

degradationthroughpublicandprivatepartnerships.

InitiallywaspresentedbyExecutiveDecreeNo.33487‐MPandestablishaPresidential

Commissionwith25membersandaHighLevelCommittee.ItisascribedtotheMinistryforthePresidencyandmandatespublicorganizationstotakeactionstoprotecttheenvironment.

Advocateprocessesforlanduseplanning

accountingforpriorityareasforecosystemservicesprovision.

Promotereforestationandincreaseinprotected

areas.

Supportnationalprogramstofundnative

forests.

SupporttheformulationandimplementationoftheNationalStrategyonClimateChange(ENCC).

Promotedisseminationofclimatechangepolicies

andstudiestolargepublicandrelevantorganizations.

Fosterinitiativesto

158

protectEcosystemServicesinlanduseplanningfortheirimportancefor

developmentandadaptationtoclimate

change. National National

Institutefor

Meteorology(IMN)

ScientificandtechnicalInstituteascribedto

MINAE.

1988CreationofIMN

Coordinatemeteorologicaldatacollectionanalysisand

dissemination.

LawN°5222createdtheIMNunderMAG.Afterwards,LawNº7152in1990transfersIMN

toMINAE.

Implementresearchonclimatechangeandvariability,andin

collaborationwithotherorganizationsstudiesonimpactsandvulnerabilityofwaterresourcesandof

coastal,agricultural,biodiversityandhealth

sectors.

Coordinatetherecompilationofthe

NationalCommunicationandispartoftheNationalDelegationtoUNFCCC

Nationaltolocal

NationalInstitute

forTechnological

Innovationin

Agriculture(INTA)

Promoteresearchonthefieldforinnovationin

agriculture

In1999theformerTechnicalResearchofficeofMAGistransformedinINTA

DisseminateresultstogetherwithMAG

decentralizedextensionoffices.

Validateappropriatesoilmanagementtechniques

topreventerosion

PersonalInterviews

National Agronomic

ResearchCentreofthe

Universit

Implementbasicandappliedresearchonsoilsciencealsothroughallianceswithlocal

producers’associationsoragriculturalextension

FormalAgreementwithMAG,1dec.1950;FormalcreationbyUCRfacultydecree,1971

159

yofCostaRica(CIA‐UCR)

offices.

Disseminateresearchresultsthroughtechnicalpamphletsorscientificarticlesinnationalandinternationaljournals.

Developandupdatedatabasesofinformationonnaturalresourcesin

CostaRica.

Providetechnicalassistanceforspecific

environmentalmanagementproblemsrelatedtoagriculture.

National Programfor

NationalStrategy

onClimateChange(ENCC)

NationalinitiativeledbyMINAEinsideIPNwithamandatetorespondtoglobalclimatechange

throughnationalprivateandpublicpartnerships.

Website(http://www.encc.go.cr/)andinterviews.

PromotepoliciestomitigateGHGwith

nationalactivities(bio‐fuels,increaseenergyuseefficiency,promotelowcarbonenergysuchashydro,wind,solar,andgeo‐thermalpower).

Promoteadaptationtoclimatechangeandvariabilityinwater,agricultural,coastal,

health,biodiversityandinfrastructurepolicies.

Supportinitiativestoraisesocialandhumanresourcestoreduce

climatechangeanditsimpacts.

Promoteawarenessincivilservantsandpublicingeneralonclimatechange

anditsimpactsandsolutions.

Identifytheneedforinformationnetworks

creationinordertofoster

160

adaptation. National Program

forIntegrat

edEnvironmentalManagementofthe

UniversityofCostaRica

(PROGAI‐UCR)

2005thePROGAIisestablishedinUCR

Promoteresearchonappropriatemanagement

ofecosystemswithspecialemphasisonwater

resources.

12‐15Feb2008firstNationalCongressPROGAI‐UCRfinaldeclarationstatesimportanceof

preventiveecosystemmanagementtorevertdegradationandfosterdevelopmentandthatestablishingpartnershipsisimportantfor

fosteringsustainablewatershedmanagement Promotenationalevents

fordisseminationofresearchresultsonwater

andwatershedmanagement.

Promotenationalprocessesfordesignofwaterrelatedpolicies.

Watershed ASA‐PacayasMinistry

ofAgricultu

re(MAG)

LocalextensionofficeoftheMAGpromotingdiffusionoftechnicalinformationonsoil

management,takesparttoregionalcommitteesto

reduceimpactsofclimate‐extremeson

agriculture.

The2004XIVSummitofIberoAmericanGovernmentssupportedthecreationofruralcentretosupporttrainingandpromotionofsustainableagriculturalproductionsystems

Promoteandsupportscollaborationswithresearchprojectson

strategiestoimprovesoilmanagementandprovidestransferto

farmers.

Decree31623MAG‐H,2003;definingprioritiesandexecutionoffundingfortechnical

assistance

Promoteawarenesscampaignsonenvironmentaldegradation.

Establishallianceswithindividualfarmersand/orfarmers'associationsto

promotesoil

161

Establishallianceswithindividualfarmersand/orfarmers'associationsto

promotesoilmanagementinthe

watershed.

National NationalAdvisoryCommissiononDegradedLands(CADETI)

Inter‐organizationalcommissionunderMINAEandMAGintegratedbyrepresentativesfrom

academicorganizations,NGOs,MINAE,IMN,MAGtechnicalconservation

offices.

ExecutiveDecreeNº27258‐MINAEcreatesCADETIasfocalpointtoUNCCDmandatingthedesignofrelatedactionsatnationallevel,andtheelaborationofNationalPlantoCombat

LandDegradationandDesertification.

FormulateandevaluatetheNationalactionplanonDegradedlands.

AdviseNationalFocalPointtoUnitedNations

ConventionandcoordinatestheNationalActionPlanstoCombatDesertification(UNCCD)

163

Annex 2. Ecosystem-based adaptation to climate change: what role for policy-makers, society and scientists?

Paper published: Vignola, R., Locatelli, B., Martinez, C., Imbach, I., 2009. Ecosystem‐Based

adaptationtoclimatechange:whatroleforpolicy‐makers,societyandscientists?.Mitigationand

AdaptationStrategiesforGlobalChange14,601‐696

Abstract

Indevelopingcountrieswhereeconomiesandlivelihoodsdependlargelyonecosystemservices,

policies foradaptationtoclimatechangeshould take intoaccountthe roleoftheseservices in

increasingtheresilienceofsociety.Thisecosystem‐basedapproachtoadaptationwasthefocus

ofaninternationalworkshopon"AdaptationtoClimateChange:theroleofEcosystemServices"

heldinNovember2008inCostaRica.Thisarticlepresentsthekeymessagesfromtheworkshop.

Keywords:Adaptationtoclimatechange,ecosystemservices,policy‐science‐societydialogue.

A2.1Introduction

An internationalworkshopon “Adaptation to Climate Change: the role of Ecosystem Services”

washeldinNovember3‐5,2008intheInternationalCentreforTropicalAgricultureResearchand

Higher Education (CATIE) in Turrialba, Costa Rica. The 80 participants came from 24 countries

(mostly Latin American) and 56 institutions (local communities, NGOs, scientific organizations,

government agencies, public and private firms, and international agencies for cooperation) to

discusstheroleofecosystemservicesinadaptationtoclimatechange.

Whilemostdevelopingcountries’economiesdependlargelyonnaturalresourcesandecosystem

services,thecapacityofecosystemstoprovideservicestosectorsofthesocietyispressuredby

landusechangeandclimatechange.Asadaptationisneededforthesectorsofsocietyfacingthe

impactsofclimatechange,itisnecessarytoconsiderthevulnerabilityandroleofecosystemsin

sustainedprovisionofecosystemservicestothesesectors.

Theaimofthisarticleistopresentthekeymessagesthattheparticipantsdiscussedduringthe

workshop.Thesekeymessagesaretargetedatthestakeholdersthatcandesignandimplement

ecosystem‐based adaptation (EBA). We define ecosystem‐based adaptation as the adaptation

policies and measures that take into account the role of ecosystem services in reducing the

vulnerability of society to climate change, in a multi‐sectoral and multi‐scale approach. EBA

164

involvesnationalandregionalgovernments, localcommunities,privatecompaniesandNGOsin

addressingthedifferentpressuresonecosystemservices,includinglandusechangeandclimate

change,andmanagingecosystemsto increasetheresilenceofpeopleandeconomicsectorsto

climate change. The concept of EBA has emerged recently in the international climate change

arena,withcountries(e.g.,Colombia,SriLanka),groupsofcountries(e.g.,theAfricanGroup)and

observers (e.g. the International Union for Conservation of Nature) addressing EBA in their

submissionstotheUnitedNationsFrameworkConventiononClimateChange.

A2.2Contentoftheworkshop

Theworkshop included seven sessions.The first sessionaimedatupdatingparticipantson the

mainconceptsandissuesrelatedtoecosystemservices,adaptationtoclimatechange,andEBA.

The second session aimed at communicating results of scientific studies related to the

vulnerabilityofecosystemsandecosystemservicesinLatinAmerica.Discussiontopicsaddressed

differentdimensionsofecosystemvulnerability:exposure(climatechangescenarios),sensitivity

(e.g.theimpactsofclimatechangeonprotectedareas,hydrologicalecosystemservicesorforest

fires), and adaptive capacity (e.g. the role of landscape connectivity in ecosystemadaptation).

Thethirdsessionaddressedtheroleofecosystemservices(e.g.hydrologicalservicesforsectors

dependingonwater,suchasdrinkingwaterandhydropowerproduction).

During the fourth session, attendees reviewed experiences of policies and measures for

adaptationatlocal,nationalandregionallevels.Eventhoughtheimpactsofclimatechangeand

theroleofecosystemservicesareuncertain,no‐regretpolicyresponsescanbenefitbothclimate

changeadaptationandsustainableecosystemmanagement.Thefifthsessionaimedatexploring

communication strategies regarding adaptation to climate change. Establishing an efficient

dialogue between scientists, decisionmakers and civil society is challenging but necessary for

EBA. This dialogue should start at the stage of research design and be pursued until the

communicationofresearchresultsanduncertainties.

During the sixth session, financial mechanisms for adaptation or for the management of

ecosystem services were reviewed. Innovative financial mechanisms are needed for these

processes,butremainchallengingbecauseofthecomplexityofevaluatingadaptationcostsand

benefits,andthesensitivityofpoliticalnegotiationsrelatedtointernationaladaptationfinance.

165

The seventh session was aimed at sharing experiences of public and private initiatives of

ecosystemmanagementandconservation.Asthese initiativesare seldomexplicitly linkedwith

climate change adaptation, their role in reducing the vulnerability of ecosystems and society

needstobefurtheranalyzed.

The participants presented awide array of perspectives and experiences related to ecosystem

servicesandadaptationtoclimatechange inLatinAmerica, includingcommunitymanagement,

paymentforecosystemservices,andinsuranceschemes.Theseexperiencesallowedparticipants

to discuss crucial issues such as the need for institutional changes for adaptation, a better

understanding of uncertainties and strategies to copewith them, improved dialogue between

scientistsanddecisionmakers,andstrengthenedlinkagesbetweensectorsandscales,fromlocal

togloballevels.

At the end of the workshop, the participants worked in groups to elaborate key messages

directedatthreegroupsofstakeholdersinEBA:(1)nationalpolicy‐makers(2)localcommunities,

privatesectorandothersmembersofcivilsociety,(3)scientists(Figure1).

Figure 1: Key messages to stakeholders related to ecosystem-based adaptation (EBA)

166

A2.3Keymessagesfornationalpolicy‐makers

Mainstreamadaptationandecosystemservices intonationalpolicies. Ecosystemdegradation

and vulnerability to climate change are development issues rather than strictly environmental

problems. As the loss of natural capital and the associated vulnerabilities are a threat for

sustainable development, national development policies should integrate ecosystem

managementandadaptation to climatechange.Multi‐sectorial andcross scaleapproachesare

needed formainstreamingbothadaptationand ecosystemservices intopolicies.Policy‐makers

should create and enforce linkages between ecosystem managers and vulnerable sectors

benefiting from ecosystem services. Moreover, education and outreach policies should raise

societal awareness about the relevance of ecosystem services and adaptation for sustainable

development.

Develop innovative funding. Because of market failures, current regulations fail to conserve

ecosystem services that are valuable for society. Thus, sectors benefiting from ecosystem

services should be involved in funding ecosystem management or conservation. Payment for

EcosystemServices(PES)isaninnovativemechanismbeingdevelopedinseveralLatinAmerican

countries in both governmental and private initiatives. PES can complement international

adaptationfundingsourcesincaseswheretheusersofecosystemservicesarewillingandableto

provide resources forprotecting these services. Thechallenge forpolicymakers is to createan

institutionalenvironmentthatfacilitatesagreementsbetweenusersandprovidersofecosystem

services.

Influence internationalpolicies. Internationalnegotiationsondevelopmentandenvironmental

issuescaninfluencenationalpoliciesonadaptationandecosystemsandthusEBAdesign.Policy‐

makersshouldthereforeensurethattheoutcomesofnegotiationsinthisarenaincludeadequate

consideration of ecosystem services that are priorities for their national development and

adaptation. To enforce their positions in these negotiations, it is strategically important that

policy‐makers strengthen their understanding of the challenges implied in designing and

implementingEBA.Thismightbeespeciallyrelevant inthecontextofthenegotiationsdefining

thetermsofreferenceandthemanagementoftheAdaptationFundwhereinvestmentpriorities

aredefined.

167

Strengthen the links between adaptation andmitigation.Mitigation instruments, such as the

Clean DevelopmentMechanism or the currently discussed REDD (Reduction of Emission from

DeforestationandForestDegradation)canprovidebenefitsforbothmitigationandadaptation,

astheycontributetoconservingandrestoringecosystemservices.However,someconcernshave

been raised about the potential negative impacts that somemitigation projectsmay have on

local development and biodiversity. Thus policymakers should try to foster synergies between

mitigation and adaptation by developing and applying guidelines or standards for mitigation

projects. These standards could provide guidance to mitigation project developers willing to

addressadaptationandenablepolicymakersordonorstoassessthecontributionofmitigation

projectstoadaptation.

Interactwithlocalcommunities.Localcommunitiesareimportantdecision‐makersinadaptation

and ecosystem management. National policymakers should empower local and indigenous

communitiestofacilitateadaptationprocessesthattaketraditionalknowledgeintoaccount.EBA

policiesshouldrecognizethediversityoflocalsituationsandcreateafacilitatingenvironmentfor

effective local adaptation and ecosystem management. Policies should also promote

environmentaleducationforpromotingEBAinlocalcommunities.

Interactwith scientists. Understanding climate change impacts and vulnerability is a complex

task, and technical expertise is required to understand and interpret the results of scientific

studies.Itisthereforeimportantthat,forbuildingno‐regretpolicies,policy‐makersinteractwith

scientists to take into account uncertainties inherent in climate change studies. Policy‐makers

shouldcreateinstitutionalarrangementsandfundingsourcestofacilitatethedevelopmentand

useofrelevantsocialandnaturalscienceknowledgeintopolicyprocesses.

A2.4Keymessagesforcommunities,localactorsandothersmembersofcivilsociety

The followingmessagesareaddressed to thevarietyof stakeholders thatmanageecosystems,

implementadaption,oruseecosystemservices,suchaslocalcommunities,thecivilsocietyand

theprivatesector.

Define and implement adaptation. Even though national policies influence local processes,

adaptationeventuallyhappenslocally.LocalactorshavetheresponsibilitytopromoteEBA,given

their direct interest in ecosystemhealth and provision of services. Communities should design

168

and implement strategies for ecosystem‐based adaptation as part of their local resource

planning.Civilsocietyandcommunitiesshouldincreasetheircapacitytonegotiateandestablish

equitable partnershipswithavarietyofactors (publicandprivate)actingatdifferent scalesof

decisions.NGOscanplayaroleinstrengtheningindigenouspopulationsinprotectingtheirrights

andvaluesinthedesignofadaptationplans.

Rewardecosystemserviceproviders.Localcommunities,civilsocietyandprivateactorsshould

rewardactorsthatconserveorrestoreecosystemservices.Thiswouldenablerestoringservices,

orpreventingfuturelandusechangesandlossofservices.Innovativemechanisms,suchasPES,

maybeanimportantcomponentofecosystem‐basedadaptation.

Interact with policy makers. Policy makers set the institutional context in which ecosystem‐

based adaptation can take place. The interests, obstacles and capacities of local communities,

however,arenotalwaysreflectedinnationalandlocalpolicies.Akeyroleforcivilsocietyisthen

toempowerthecapacityoflocalactorstoparticipateinpolicy‐makingatbothlocalandnational

levels.Building capacityandopening spaces forparticipation inpolicymakingcanbeaway to

achievethesegoals.

Interact with scientists. Research on ecosystems services and how they can reduce societal

vulnerability to climate change is a relatively new field characterized by complexity and

uncertainty. In this context, local knowledge on ecosystems and theirmanagement should be

respected and is crucial for improving scientists’ understanding and their capacity to support

design of appropriate responses. In this respect, interactionwith social scientists is important

giventheirexpertiseonintegratingdifferentknowledge‐systemsintocollaborativeefforts.Local

communitiesandcivilsocietyarethuscalledtoplayamoreactiverole,gettinginvolvedinfield

researchandinformingscientistsonobservedchangesandlocaladaptation,andscientistsshould

adapt their methods to use these resources. In addition, private sectors could also finance

science,astheyalsocouldbenefit fromnewscientificknowledgeonadaptationandecosystem

services.

A2.5Keymessagesforscientists

Quantify and value ecosystem services. Several international scientific initiatives such as the

MillenniumEcosystemAssessmenthaveacknowledgedtherelevanceofecosystemservices for

169

human wellbeing. However, more evidence is needed on the role of ecosystem services in

reducingthevulnerabilityofsociety,aswellasthecostsandbenefitsofconservingecosystem

services,inthecontextofclimatechange.Quantificationandspatialprioritizationofecosystem

servicesforadaptationarealsoneededforadaptation.

Evaluate uncertainties. Ecosystem services and climate change impacts assessments dealwith

complexityanduncertainty;therefore,theirrecognitionandevaluationisessentialfordesigning

adaptiveapproachestoEBA.Impactorvulnerabilitystudiesshouldalwaysincludeananalysisof

sensitivity, for instance, using differentmodels and different climate scenarios. In an adaptive

process,uncertaintyanalysisalsoenablestheidentificationofwhichvariablesareuncertainand

determinant in explaining vulnerability to climate change. These variables can be transformed

intoindicatorsformonitoringtrendsinthefield.

Work at local scales. Many climate change impact studies or vulnerability assessments show

resultsonregionalandglobalscales,providingabroadviewthatseldomhelpsinterestedparties

todesignlocaladaptation.Forinstance,EBArequiresidentifyingtheflowofecosystemservices

in local landscapes, as a tool for determining land uses that may provide ecosystem services

relevanttotheadaptationofsociety.Thus,moreresearchisneededtoaddressscientificissues

related to EBA at a local scale. In addition, social scientists have an important role to play in

understandinglocalsocialprocessesrelevantforadaptationandecosystemconservation.

Communicate results to non‐scientists. Scientists should devote substantial efforts in

communicatingresearchresultstonon‐scientists,suchaslocalNGOs,publicadministrationand

themedia, inordertoincreasetheircapacityto influencethe implementationofEBA.Strategic

allianceswithexpertsfromcommunicationsciencesshouldbeconsideredasakeycomponentof

researchprojectsonEBA.

Interactwith localcommunitiesandprivatesector.Scientistsstudyingclimatechange impacts

andthedesignofEBAmeasuresarecalledtoworkcloselywithlocalcommunitiesandtheprivate

sector in order to understand local processes of adaptation and ecosystem management.

Scientists should also involve local communities and private sectors from the early stages of

research design, so that scientific work and results can serve local EBA design and

implementation.

170

Interactwithpolicymakers.ScientistsproducingknowledgeonclimatechangeimpactsandEBA

measures should ensure that results are communicated in a way that is relevant to policy‐

makers.Atthelocalandnationallevels,scientistsshouldparticipate,asoneofthestakeholders,

inpolicydesignprocess,includingproblemidentification,strategyformulation,selectionofpolicy

options,monitoringandevaluation.Atgloballevel,scientistscanplayanimportantadvisoryrole

tonationalpolicy‐makerswhoparticipateininternationalnegotiationsrelatedtoadaptationand

possiblytoEBA.

A2.6Conclusions

Increasing attention has been paid to the complex interactions between human and environmental

systems,especially regarding theconservationandmanagementofecosystemservices.However, the

roleofecosystemservices in supportingadaptation to climatechange isa relativelynew issue in the

scientific arena, and even more so in the policy arena. In order to promote the role of ecosystem

servicesinsocietaladaptationtoclimatechange,itisimportantthatavarietyofactors,suchaspolicy‐

makers, scientists and civil society, are actively involved in accordancewith theirmandates. The key

messagespresented inthisarticlearenotonly relevant forLatinAmericabutarevalidalsoforother

regions,especiallywherepeopledependstronglyonnaturalresources.

172

Acknowledgments

In long journey of this thesis I havemet and collaboratedwithmany people that have shared

withmetheirknowledgeandfriendship,andhavecontributedtomyprofessionalandpersonal

growth.Firstofall,IwanttoacknowledgethesupportgiventojointhePhDprogramatETHZby

Dr. LucioPedroni, leaderof theClimateChangeProgramatCATIEat themomentofPhDstart

date. He has given me the opportunity to participate in the TroFCCA project where my

understanding of global change phenomena has been fed by intense exchange of theories,

knowledge and methods that have inspired my research in the PhD. I want to thank Bruno

LocatelliwhomduringtheinitialyearsattheCATIEhassharedhisknowledgeandthefriendship

of hiswarm family. I acknowledge Pablo Imbach, a friend from the very start and a colleague

sharingmotivationsandperspectivesonprofessionalaspectsaswellasthepositiveandnegative

momentsofourPhDstudies. IacknowledgeprofessorR.W.Scholz forthe inspiringdiscussions

hadduring ourmeetings. Forhis supportand foropening the opportunity to joinhisgroupat

ETHZwhereImetpeoplethathavebeenfriendsandteacherstomewhilesharingmethodsand

theory of interdisciplinary science, so needed especially in developingworld contexts. Iwould

alsoliketoacknowledgeMariaReyforherkindsupportthroughoutmystudiesatETHZ.Ithank

ProfessorThomasKoellnerforhisfriendshipandsupportinmanyjourneystoZurich.Ithankhim

for his support in the developing of ideas and opportunity for this PhD; from the very start

Thomashasbeenanimportantmotivatoralongtheupsanddownsofthisresearch.

I want to give special thanks to Professor Tim L. McDaniels from the University of British

Columbiawhom kindly acceptedmy invitation to visit CATIE and join the research developed

during this investigation. Timhas givenmehis valuable friendship and has been an important

inspirerforthisthesisandforexpandingourscientificandpersonalcollaborationandfriendship.

IthankallmyfriendsandcolleaguesattheClimateChangeProgramatCATIEthatsharethedaily

stressesandpleasuresofourjobswithsmilesandpatience.

IwanttothankalsomyfamilyinItalyforalwaysbeingthereandmakingtheAtlanticOceanand

theMediterraneanSeadisappearwhenwetalk, andthedaysandmonthsseparatingourvisits

disappearwhenwemeet.Finally,IwanttothankthetwotreasuresofmylifeSofiaManuelaand

Jimenawhommakemylifeacontinuousmotivationtoimprovemycharacterandaninspiration

togivemoretoothers.

174

Curriculum Vitae

RaffaeleVignola

Agr.Eng.UniversityofFlorence,Italy.

1969 BorninSalerno,Italy

1982‐1987 ScientificLyceum,Eboli,Salerno,Italy

1987‐1996 StudyatUniversityofAgriculture,Florence,Italy

1994 Field‐studiesinEcologyTamilNadu,India

1997ScholarshipAgro‐MeteorologicalInstitute,UniversityofFlorence,Italy

1998FellowshipatInstitutefortheOverseas,MinistryofForeignAffairs,Italy

2000ConsultantonWatershedManagementPlanning,Honduras

2001‐2003CoordinatorEUDisasterPreventionProjectDIPECHOII,ElSalvador

2003‐2004StudyMasterofScienceinEnvironmentalEconomics,CentreforHigherEducationandTrainingforLatinAmerica(CATIE),CostaRica

2004UnitedNationsOperationProgram(UNOPS)‐CentralAmericanProgramtoFightPoverty,ConsultantSocio‐economicimpactsofenvironmentalpollution,ElSalvador

2005‐2010Researcher/professorofClimateChangeProgramatCATIE,CostaRica

2007‐2010 PhDstudiesatETH‐NSSIwithCATIE

2010‐2012Co‐investigatorresearchproject:HumanDimensionofSoilRegulationServicesunderclimatechangeinCostaRica,ClimateChangeprogram‐CATIE/InstituteforResourcesandSustainability‐UniversityofBritishColumbia,Canada

Contact

ClimateChangeProgram,

TropicalAgricultureCentreforHigherEducationandTraining(CATIE)

7170,Turrialba,

CostaRica

phone:+506‐25582528,email:[email protected];[email protected];www.catie.ac.cr

Rights / License: Research Collection In Copyright - …1995/eth... · Prof. Dr. Roland W. Scholz,...

Documents

Transcript of Rights / License: Research Collection In Copyright - …1995/eth... · Prof. Dr. Roland W. Scholz,...