These Mlvr
These Mlvr
These Mlvr
THÈSE
PRÉSENTÉE
PAR
OCTOBRE 2013
Composition du jury :
Avertissement
La diffusion de ce mémoire ou de cette thèse se fait dans le respect des droits de son auteur,
qui a signé le formulaire « Autorisation de reproduire et de diffuser un rapport, un mémoire
ou une thèse ». En signant ce formulaire, l’auteur concède à l’Université du Québec à
Rimouski une licence non exclusive d’utilisation et de publication de la totalité ou d’une
partie importante de son travail de recherche à des fins pédagogiques et non commerciales.
Plus précisément, l’auteur autorise l’Université du Québec à Rimouski à reproduire, diffuser,
prêter, distribuer ou vendre des copies de son travail de recherche à des fins non
commerciales sur quelque support que ce soit, y compris l’Internet. Cette licence et cette
autorisation n’entraînent pas une renonciation de la part de l’auteur à ses droits moraux ni à
ses droits de propriété intellectuelle. Sauf entente contraire, l’auteur conserve la liberté de
diffuser et de commercialiser ou non ce travail dont il possède un exemplaire.
REMERCIEMENTS
Je remercie aussi celui qui m’a appris que la vie est une fête remplie d’amour et celle
qui m’a appris que la vie est constituée de spiritualité et de responsabilité, je remercie tous
ceux qui ont été près de moi et avec moi – famille et amis – lors des blessures et dans les
combats, avec un respect constant à l’égard de ma personne.
AVANT-PROPOS
Cette recherche propose des solutions aux problèmes rencontrés par les projets éoliens
au Québec sur les aspects d’acceptabilité sociale. Lors de l'élaboration ou de la mise en
fonction de certains projets éoliens, une partie de la population a demandé une plus grande
implication dans la prise de décisions et la prise en compte de ses préoccupations. La prise de
décision basée sur une analyse multicritère et multiacteurs a connu un développement
accéléré partout dans le monde et a été utilisée dans de nombreux secteurs économiques.
C'est pour cette raison que cette recherche a été réalisée en collaboration entre le Laboratoire
de recherche en énergie éolienne de l’Université du Québec à Rimouski et le département de
géographie de l’UQAM, par l’intermédiaire du GEIGER et du GERAD. Il s'agit d'un travail
indépendant qui tient compte de façon égale des intérêts et points de vue des acteurs
impliqués ou affectés par le développement des projets éoliens.
Pour mener à terme cette recherche, nous avons également collaboré avec Mme Marie-
Josée Fortin, professeure à l'UQAR, experte des impacts sur le paysage, une des
préoccupations principales exprimées au sujet des projets éoliens en Gaspésie. La grille
d’analyse MOFF a été envisagée lors de la conférence CanWEA 2010, en écoutant une
conférence de Richard Delaney. C’est dans ce contexte interdisciplinaire que l’approche
DEMIT se nourrit pour répondre aux questions du développement durable des projets éoliens.
Cette recherche se veut un outil permettant d'améliorer les scénarios de projets éoliens
du point de vue de l'ensemble des acteurs. Des échanges avec des experts chevronnés ont
nourri l'élaboration de l’approche qui a été validée auprès des vrais acteurs de la mise en
place d’un parc éolien.
TABLE DES MATIÈRES
CHAPITRE I
INTRODUCTION GÉNÉRALE .............................................................................................. 1
1.1 Problématique ................................................................................................................. 1
1.2 Objectifs.......................................................................................................................... 4
1.3 Méthodologie générale ................................................................................................... 6
1.4 Plan ............................................................................................................................... 11
CHAPITRE II
COUPLING MCDA AND GIS IN A DECISION MAKING PROCESS FOR WIND FARM
PROJECTS ANALYSIS –TIMED APPROACH ................................................................... 13
2.1 Introduction .................................................................................................................. 14
2.2 Theoretical framework.................................................................................................. 16
2.2.1 Concept definition (visual information, model, scientific knowledge/local knowledge,
wind energy project, MCDA, scenarios, GIS, stakeholders) ........................................ 16
2.2.2 Participatory GIS is a source of transparency during scenario modeling ..................... 20
2.2.3 MCDA-GIS coupling as a source of negotiation: limitations and possibilities ............ 20
2.2.4 The MCDA-GIS coupling as a sustainable development approach ............................. 26
2.3 The MCDA-GIS multi-stakeholder coupling as a source of transparency, consensus
and possibilities of sustainable development ................................................................ 28
2.4 The TIMED approach as a negotiation facilitation instrument .................................... 30
2.4.1 Presentation of the TIMED approach as a negotiation facilitator................................. 30
2.4.2 MCDA module ............................................................................................................. 31
2.4.3 GIS Module .................................................................................................................. 34
viii
CHAPITRE III
TERRITORIAL INTELLIGENCE MODELING FOR ENERGY DEVELOPMENT
(TIMED) – A CASE STUDY FOR THE BAIE-DES-SABLES (CANADA) WIND FARM
............................................................................................................................................... .53
3.1 Introduction .................................................................................................................. 54
3.1.1 Overview of wind power development in Québec, Canada and Baie-des-Sables wind
farm............................................................................................................................... 54
3.1.2 The TIMED approach ................................................................................................... 55
3.1.3 Postulates for the elicitation of scenarios ..................................................................... 58
3.2 Implementation steps .................................................................................................... 59
3.2.1 MCDA Module ............................................................................................................. 59
3.2.2 Participatory GIS Module ............................................................................................. 65
3.2.3 Contributive Involvement of Stakeholder (CIS) Module ............................................. 68
3.2.4 Scientific Knowledge and Local Knowledge (SK-LK) Module.................................. .68
3.3 Results .......................................................................................................................... 69
3.4 Discussion ..................................................................................................................... 73
3.5 Conclusion .................................................................................................................... 73
CHAPITRE IV
MCDA: MEASURING ROBUSTNESS AS A TOOL TO ADDRESS STRATEGIC WIND
FARMS ISSUES ..................................................................................................................... 78
4.1 Introduction .................................................................................................................. 79
4.1.1 Four TIMED Approach Modules: Literature Review .................................................. 80
4.1.2 An extended concept of robustness .............................................................................. 82
4.1.3 Broadened robustness as an answer to the modeling of strategic concerns .................. 83
4.1.4 Understanding the TIMED approach to understand the construction of matrices subject
to the robustness analysis.............................................................................................. 84
4.2 Robustness analysis of the strategic concerns: a simulation to address the robustness
concerns ........................................................................................................................ 87
4.2.1 Strategic level postulates .............................................................................................. 87
4.2.2 Methodology of the robustness analysis of strategic level postulates .......................... 88
ix
CHAPITRE V
IDENTIFIER LES FACTEURS DE RÉUSSITE D’UNE DÉCISION PARTICIPATIVE PAR
L’APPROCHE DEMIT : UNE ANALYSE FFOC .............................................................. 111
5.1 Introduction ................................................................................................................ 112
CHAPITRE VI
CONCLUSION GÉNÉRALE ............................................................................................... 142
6.1 Contributions .............................................................................................................. 142
6.2 Applications ................................................................................................................ 144
6.3 Limites et extensions de la recherche ......................................................................... 145
6.4 Mot de la fin…………………………………...……………………….…..………148
ANNEXE A
PRÉCISIONS MÉTHODOLOGIQUES .............................................................................. 151
ANNEXE B
COMPLÉMENT DE RÉSULTATS ..................................................................................... 163
ANNEXE C
CONSTRUCTION DES CRITÈRES ................................................................................... 195
ANNEXE D
CARTES THÉMATIQUES .................................................................................................. 208
ANNEXE E
CALCULS SUR EXCEL POUR CONSTRUIRE LES ÉCHELLES DE MESURE DES
INDICATEURS .................................................................................................................... 216
ANNEXE F
16 FEUILLES D’ANALYSE MOFF ................................................................................... 224
ANNEXE G
LISTE DÉTAILLÉE DES FORCES-FAIBLESSES ET DES OPPORTUNITÉS-
CONTRAINTES IDENTIFIÉES LORS DE L’ANALYSE MOFF ..................................... 247
BIBLIOGRAPHIE……………………………………………….........................................255
Introduction générale et conclusion générale………………………………….…..………..255
Annexes……………………………………………………………………….…………….258
LISTE DES FIGURES
Figure 4.5 Ranking of scenarios using the combined Nmax-Npro evaluation matrix of the
Type of ownership of wind farm-Public-private partnership postulate .................................. 98
Figure 4.6 Rules followed for the changes in the scenario rankings…………..……………107
Figure 5.1 Histogramme du nombre de forces, faiblesses, opportunités et contraintes de
DEMIT intervenant dans les espaces nécessaires à la décision participative. ...................... 125
Figure 5.2 Histogramme du nombre de forces, faiblesses, oppotunités et contraintes des
modules SIG et AMCD de DEMIT ...................................................................................... 126
Figure 5.3 Graphique radar du nombre de forces-faiblesses et opportunités-contraintes par
étape de l'approche DEMIT .................................................................................................. 127
Figure 5.4 Graphique radar du nombre de faiblesses par étape de l'approche DEMIT ..........
......................................................................................................................... 128
Figure 5.5 Graphique radar du nombre de contraintes par étape de l'approche DEMIT ........
......................................................................................................................... 129
Table 3.1 Set of criteria for the analysis of wind farm ............................................. .61
Table 4.1 Models and matrices requised according to the different scenarios,
postulates and power weighting options considered in the robustness analysis............89
Table 4.4 Evaluation matrix with different grades for the Communication tools –
Negotiation table postulates...........................................................................................95
Table 4.6 Combined evaluation matrix for the grading of the postulate Type of
ownership of wind farm – Public – private partnership ............................................... 98
Tableau B.1 Critères proposés par DEMIT, classifiés selon les dimensions du
développement durable. .............................................................................................. 172
Tableau B.6 Construction des échelons pour évaluer le critère 9 « impact négatif
potentiel sur le tourisme.............................................................................................. 183
Tableau B.7 Construction des échelons pour évaluer le critère 10 « emplois ». ..... 184
Tableau B.8 Construction des échelons pour évaluer le critère 11 « avantages socio-
économiques régionaux. ............................................................................................. 184
Tableau C.3 Regroupement des préoccupations sur la prise en compte des valeurs
et des besoins des populations……………………………………………………….198
US United States
TWh Térawatt-heure
kW Kilowatt
kWh Kilowatt-heure
MW Mégawatt
dB Décibel
RÉSUMÉ
DEMIT est représenté visuellement par une figure exprimant l’idée de co-construction
des scénarios pour la prise de décision avec tous les acteurs impliqués qui se retrouvent au
centre de cette approche. DEMIT est composé de quatre modules : aide multicritère à la
décision, implication contributive des acteurs, systèmes d’information géographique
participatifs, et connaissances scientifiques-connaissances locales. L’intégration de ces quatre
modules permet d’analyser différents scénarios d’implantation des éoliennes tout en incluant
les préoccupations de tous acteurs impliqués par le projet. Cela conduit, grâce au processus
de prise de décision proposé, au choix du meilleur scénario collectif. Le deuxième objectif a
permis de tester DEMIT dans le cas de l’évaluation ex post d’un parc éolien en opération
depuis 2006. Onze personnes y ont participé, représentant quatre catégories des acteurs : le
xxiv
secteur privé, le secteur public, les experts et la société civile. Ce test nous a permis
d’analyser le contexte actuel dans lequel les projets éoliens se développent au Québec. Les
préoccupations de certains acteurs sortant du contexte de l'étude de cas ont été explorées dans
le troisième volet. Celui-ci nous a permis de faire des simulations en tenant compte des
postulats de niveau stratégique, tel que "le type d’outil de communication" utilisé pour
approcher la communauté hôte du projet et "le type de propriété" du parc éolien. Finalement,
le quatrième volet, portant sur l'analyse FFOC et réalisé avec la participation de huit experts,
nous a permis de vérifier dans quelle mesure DEMIT réussit à construire quatre espaces
garants de la prise de décision participative, à savoir : les espaces physique, intellectuel,
émotionnel et procédural. De cette analyse, 116 forces, 28 faiblesses, 32 contraintes et 54
opportunités ont été répertoriées.
The TIMED approach is visually represented by a figure expressing the idea of a co-
construction decision and where all stakeholders are the focus of this methodology. TIMED
is composed of four modules: Multi-Criteria decision analysis, participatory geographic
Information systems, active involvement of the stakeholders and scientific knowledge/local
knowledge. The integration of these four modules allows for the analysis of different
implementation scenarios of wind turbines in order to choose the best one based on a
participatory and transparent decision-making process that takes into account stakeholders’
concerns. The second objective enabled the testingof TIMED in an ex-post experience of a
wind farm in operation since 2006. In this test, 11 people participated representing four
stakeholder' categories: the private sector, the public sector, experts and civil society. This
test allowed us to analyze the current situation in which wind projects are currently
developed in Quebec. The concerns of some stakeholders regarding situations that are not
considered in the current context were explored through a third goal. This third objective
allowed us to make simulations taking into account the assumptions of strategic levels.
Examples of the strategic level are the communication tools used to approach the host
community and the park property type. Finally, the fourth objective, a SWOT analysis with
the participation of eight experts, allowed us to verify the extent to which TIMED approach
succeeded in constructing four fields for participatory decision-making: physical, intellectual,
emotional and procedural. From these facts, 116 strengths, 28 weaknesses, 32 constraints and
54 opportunities were identified.
methodology; taking into account the physical, intellectual, emotional and procedural spaces
to articulate a participatory decision; using the proposed methodology in renewable energy
sources other than wind; the need to an interdisciplinary team for the methodology
application; access to quality data; access to information technologies; the right to public
participation; the neutrality of experts; the relationships between experts and non-experts;
cultural constraints; improvement of designed indicators; the implementation of a Web
platform for participatory decision-making and writing a manual on the use of the developed
methodology.
INTRODUCTION GÉNÉRALE
1.1 Problématique
La filière éolienne a connu un développement très accéléré partout dans le monde avec
un taux annuel de croissance de l'ordre de 25% pendant les 10 dernières années. Le Québec
n'a pas fait exception et les parcs éoliens ont été installés à un rythme rapide depuis le
premier parc installé en Gaspésie en 1998. La politique énergétique élaborée par le
Gouvernement du Québec en 2005 prévoit l'installation de 4 000MW de puissance éolienne
jusqu'en 2015. Si plusieurs projets ont connu un déroulement sans heurt, d'autres ont fait face
à une vive opposition d'une partie de la population. Certaines préoccupations ou craintes des
gens par rapport à la technologie éolienne ont été exprimées souvent sous la forme d’un refus
des projets. Ceci s'est traduit par des retards importants ou même par l'abandon de plusieurs
projets. Ainsi, aujourd'hui, 700MW des 4 000MW initialement prévus ont dû être
abandonnés. Un facteur important de l'acceptabilité sociale des projets éoliens est la
participation et la prise en compte des positions de tous les acteurs lors de l'élaboration et du
choix des scénarios d'implantation des parcs éoliens. Afin de cerner l'ensemble des facteurs
qui peuvent soutenir une démarche rigoureuse de prise de décision qui soit transparente et
participative, et qui favorise le développement durable de l’énergie éolienne, les aspects
suivants doivent être pris en compte:
2
De nombreuses difficultés existent pour la réalisation des projets éoliens: les coûts de
financement semblent être plus élevés que ceux des ressources conventionnelles,
l’acceptation sociale devient une condition pour l’exécution de ces projets et le raccordement
au réseau électrique doit compter sur des règlements clairs selon les différentes tailles
d’éoliennes (domestiques : 25 m et 10 kW ; moyennes : 60 m et 500 kW ; grandes : 150 m et
3 MW, selon Gipe (2007). Ces règlements évoluent depuis 30 ans. Cependant, certaines
compagnies d’électricité demandent encore des améliorations (Gipe, 2007). Les promoteurs
des projets font face à différentes exigences techniques, telles que la sélection du site,
l’évaluation du potentiel éolien, la conception du projet ou l’élaboration d’études de
faisabilité et d’impact environnemental. Pour résoudre ces difficultés, de nombreuses études
ont été menées telles que celles sur l’introduction des énergies renouvelables dans le secteur
électrique (Tsioliaridou et al., 2005) ou celles sur les stratégies énergétiques renouvelables
pour le développement durable (Midilli et al., 2005). Dans les pays où les parcs éoliens ont
eu un grand succès, ces obstacles ont pu être surmontés. Au Québec, l’énergie éolienne est
développée depuis 1990 (Bouchard, 2007). En 2005, Québec reprend ainsi, dans sa stratégie
énergétique, son potentiel éolien technique exploitable hors des zones restrictives, lequel est
estimé à environ 3 984 000 MW (Ministère des Ressources naturelles et de la Faune du
Québec, 2005), tout en affirmant aussi que le réseau électrique d’Hydro-Québec ne peut en
admettre que 4 000 MW (Ministère des Ressources naturelles et de la Faune du Québec,
2006).
une plus grande place aux communautés locales et régionales et aux nations autochtones dans
le développement énergétique » et à « devenir un leader du développement durable ». Cette
stratégie souligne que l’exploitation du potentiel éolien est prioritaire parmi les ressources
énergétiques ciblées. La deuxième orientation de cette stratégie souligne que la filière
éolienne est rapide à développer, qu’elle possède des coûts compétitifs par rapport à d’autres
sources énergétiques et qu’elle favorise le développement durable.
Par ailleurs, la pratique a aussi montré que la prise en compte des valeurs des différents
acteurs impliqués dans l’implantation des projets éoliens, est essentielle pour leur réussite, tel
qu’illustré par le cas du parc éolien « Le Haut des Ailes » (Erelia Groupe, 2007), en France,
qui s’est développé rapidement, comparé à d’autres projets.
1.2 Objectifs
Pour atteindre cet objectif général, quatre objectifs spécifiques sont visés :
2. tester l’approche développée par une étude de cas. Ce test permettra de valider
la méthodologie résultante auprès des acteurs dans un contexte réel. Ce projet
pilote d'application de DEMIT fournira également les données nécessaires à
l'atteinte de l’objectif 3;
1. ‘Coupling MCDA and GIS in a Decision Making Process for Wind Farm
Projects Analysis –TIMED Approach’. Paru dans International of Geomatics
and Spatial Analysis, VOL 23/1 - 2013 - pp.95-123 - doi:10.3166/rig.23.95-
123. Cet article a permis la réalisation de l’objectif 1 relatif à la conception
d’une méthodologie couplant l’AMCD et les SIG participatifs.
Globalement, la recherche est appliquée et l’atteinte des objectifs suivra les quatre
étapes suivantes :
7
Une fois conçue, l’approche DEMIT a servi à analyser un parc éolien existant, celui de
Baie-des-Sables, et cela sous la forme d’une étude de cas. Ce choix a été dicté par le fait que
ce parc est caractérisé comme « un projet exemplaire à plus d’un titre » (Ressources
naturelles Canada, juillet 2008, page 25). Ce projet a été retenue par Hydro-Québec dans le
cadre du premier appel d’offres de 1000 MW éoliens lancé en 20031. Par la suite, un avis de
projet a été déposé en juin 2004 et le promoteur a réalisé l’étude d’impact sur
l’environnement que la LQE du Québec exigeait pour ce type de projet. En 2005, le ministère
de Développement durable, de l’Environnement et des Parcs a demandé au BAPE de tenir
une audience publique pour ce projet. Cette audience s’est déroulée en deux parties : la
1
http://www.mrn.gouv.qc.ca/energie/eolien/eolien-projets.jsp Page Web consultée le 10 mai 2013.
9
première les 16 et 17 mai 2005, et la seconde les 14 et 15 juin 20052. Le parc éolien a été
construit et mis en opération en novembre 20063.
Les étapes de validation lors de l'étude de cas, pour le module AMCD sont : a) la
définition du problème; b) l’identification des enjeux; c) la structuration des enjeux; d) la
sélection d’une méthode d’agrégation multicritère ; e) la sélection des critères; f) la
pondération des critères; g) la sélection des indicateurs et de leurs échelles; h) la réalisation
d’une matrice d’évaluation; i) la réalisation d’une analyse de préférence; j) la consolidation
2
http://www.bape.gouv.qc.ca/sections/rapports/publications/bape217.pdf Page Web consultée le 10 mai
2013.
3
http://www.cartierenergie.com/cartier-energie.aspx?sec=7 Page Web consultée le 10 mai 2013.
10
du jugement et k) la décision. Toutes ces étapes sont détaillées dans le chapitre III, section
3.2.1 MCDA Module. Les étapes de validation lors de l’étude de cas, pour le module SIG
sont : a) la sélection d’informations géoréférencées; b) la réalisation de la base de données; c)
la conception des scénarios à évaluer et d) la réalisation de cartes thématiques. Toutes ces
étapes sont détaillées dans le chapitre III, section 3.2.2 GIS Module. Cette étude de cas a
permis de bien cibler les moments de participation des acteurs, tel qu’indiqué dans le chapitre
III, section 3.2.3 correspondant au module ICA. Tant les connaissances scientifiques que le
savoir local ont été pris en compte dans toutes les étapes réalisées, tel qu’indiqué dans le
chapitre III, section 3.2.4, correspondant au module CS-CL. Cette étude de cas a pris en
considération le contexte actuel réel dans lequel les projets éoliens se développent au Québec.
C'est ainsi que nous avons testé, validé et ajusté l’approche DEMIT.
chapitre V, section 5.2.4.1 Implémentation d’un focus group, huit experts, divisés en deux
focus group, un pour le module AMCD et l’autre pour le module SIG, ont participé à
l’analyse FFOC. Le groupe analysant le module AMCD s’est réuni deux fois, tandis que celui
analysant le module SIG, une fois. De plus, des courriels ont été utilisés pour recueillir, en
certains cas, leurs derniers commentaires.
1.4 Plan
Cette thèse comporte six chapitres, dont quatre correspondent aux articles publiés. Le
premier chapitre (introduction) établit la pertinence de cette recherche, la problématique à
laquelle elle répond, les objectifs poursuivis, la méthodologie générale utilisée et le lien
scientifique entre les différents articles.
Dans le premier article (Chapitre II), sont présentées les caractéristiques des projets
éoliens, leur contribution pour contrer les changements climatiques, ainsi que les
préoccupations environnementales, socioculturelles et économiques exprimées par différents
acteurs sur les conséquences de ces projets au Québec. Le cadre théorique de la recherche est
basé sur deux méthodes et met en évidence les possibilités et limites pour favoriser la
transparence et l’implication dans la prise de décision et le développement durable des projets
de parcs éoliens. Le résultat de cette première étape est l’approche DEMIT qui intègre le
cadre théorique répondant à la transparence, à l’implication et au développement durable des
projets de parcs éoliens.
Le troisième article (Chapitre IV) étudie la robustesse des résultats afin de répondre à
certaines questions au sujet des projets éoliens. Différents classements des scénarios sont
12
analysés, pour arriver finalement à dégager des règles objectives d’ordonnancement des
scénarios selon le postulat assumé.
Dans le dernier chapitre (VI) nous présentons les conclusions générales tout en
identifiant les contributions, les limites et les applications futures de cette recherche.
Plusieurs éléments utilisés dans cette recherche qui n’ont pas été inclus dans les
publications à cause des restrictions d’espace, sont présentés dans les annexes : précisions
méthodologiques de l’étude de cas, résultats complémentaires de l’étude de cas, construction
des critères, cartes thématiques, calculs sur Excel pour construire les échelles de mesure des
indicateurs, feuilles de l’analyse FFOC et liste détaillée des forces-faiblesses et des
opportunités-contraintes identifiées lors de l’analyse FFOC.
CHAPITRE II
COUPLING MCDA AND GIS IN A DECISION MAKING PROCESS FOR WIND FARM
PROJECTS ANALYSIS –TIMED APPROACH
Résumé
Le développement à grande échelle de l’énergie éolienne (projets de plus de 25 MW) est une alternative
énergétique permettant la réduction des gaz à effet de serre et permet la diversification des sources d’énergie.
Cependant, au Québec, Canada, cette filière énergétique, en croissance rapide, a soulevé des préoccupations dans
la population. Parmi celles-ci on trouve la protection du paysage comme ressource du tourisme-nature. De plus, la
population hôte demande plus de transparence et implication lors de la prise de décision au sujet des projets
éoliens. Deux méthodes sont essentielles pour répondre à ces deux demandes : l’Aide à la décision multicritère
(ADMC) et les Systèmes d’information géographiques (SIG) participatifs. Le couplage ADMC-SIG a déjà été
utilisé dans différents domaines incluant le secteur éolien. Nous présentons ici une approche utilisant ces deux
4
M. de L. Vazquez ()
Ph.D student in environmental sciences at the Université du Québec à Rimouski. Wind Energy Research Laboratory.
300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
email: vazquez_rascon.maria_de_lourdes@courrier.uqam.ca
5
Ph.D., Professor, Geography Department, Groupe d’études interdisciplinaires en géographie et environnement
régional Groupe d’études et de recherche en analyse des décisions. Université du Québec à Montréal (UQAM) Case
postale 8888, succursale Centre-ville, Montréal, H3C 3P8, Qc, Canada
email: waaub.jean-philippe@uqam.ca
6
Ph.D. Professor, Mathematics, Computer Science and Engineering Department. Wind Energy Research Laboratory.
Université du Québec à Rimouski (UQAR) 300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
email: adrian_ilinca@UQAR.QC.CA
14
méthodes et qui peut être utilisée pour analyser différents scénarios d’un projet éolien à grande échelle, tout en
prenant en considération les préoccupations de tous les acteurs impliqués. Cette approche a comme but d’arriver
au meilleur scénario de parc éolien sur la base d’une prise de décision transparente et participative.
Abstract
The large-scale (more than 25 MW) development of wind energy seems to be a very promising option
given climate change and the need to diversify sources of energy. However, in Quebec, Canada, this fast growing
field is raising the concern and apprehension of the population for various reasons, the first one being the
protection of Quebec’s rural landscape, an important tourism resource. Transparency and participation in the
decision-making process are the two major requests of local stakeholders regarding the installation of wind farms.
Hence, two methods are essential to address these demands: MultiCriteria Decision Analysis (MCDA) and
Geographic Information Systems (GIS). Several scholars working in different fields, including the wind sector,
have used the MCDA-GIS coupling. Therefore, we developed an approach that use both methods specifically in
the context of large-scale wind energy projects. We emphasize how such a approach can be used in the analysis of
different scenarios and consider the opinions and attitudes of all stakeholders in order to choose the “best” one
based on a participative, transparent decision making process.
2.1 Introduction
2030 (EIA, 2009). Traditional sources of energy will continue to supply a significant part of
global energy demand. In general, during this period, most electrical energy produced from
renewable energy sources will come from hydro-electricity (54%) and wind power (33%)
(EIA, 2009). Several countries have made important efforts to set up wind energy in regions
where the exploitation of this renewable source is favourable and have seen an exponential
growth of this field (Bouchard, 2007). In fact, the wind energy sector experienced a 29%
growth in 2008 and represented more than 1.5% of worldwide electrical energy consumption
in that year, with an annual production of 260 TWh (WWEA, 2009).
In 2008, wind farms in Canada provided 1.1% of the country’s total electrical demand.
In January 2012, the wind power installed was 5,265 MW for approximately 2% of total
electricity demanded. As for September 2012, Canada’s wind farm capacity is 5,511 MW of
which 1,057 MW are in Quebec, where the exploitable wind potential is estimated at
3,984,000 MW. However, when technical limitations are taken into consideration, the
province’s national grid can currently accommodate only 4,000 MW (Hélimax Énergie Inc.
and AWS Truewind LLC 2005). The Ministère des Ressources naturelles et de la faune of
Québec has set the objective of increasing to 4,000 MW the energy generated by wind in
Quebec in 2015 (MNRF, 2010). This already represents a large-scale development in this
field.
The development of wind farms has been made through calls for tenders with the
private sector. The implementation of these farms and projects has raised concern among
local populations. In this context, the transparency of the decision-making process becomes a
priority for the stakeholders involved in the wind power section in Quebec; moreover,
because this sector will continue to develop and improve (Jegen, 2008). The impact of wind
turbines on landscapes is one of the main factors hindering the social acceptability of wind
energy projects (Wolsing, 2005). This is why participation in some decisions, such as the
selection of wind-farm sites, becomes the key to the projects’ acceptance by local
stakeholders (Wüstenhagen, Wolsing and Bürer, 2006). Social acceptability, which seems to
be the most important obstacle to the construction of wind farms, could even become a factor
reducing the growth of renewable energies as forecasted by governmental programs.
Furthermore, the social acceptability of wind farms in Quebec depends on numerous aspects,
16
including the general perception of wind energy and of Quebec’s energy policies in
particular, as well as their impact on a given region’s economic activity and the project’s
social outcomes and consequences (Jegen, 2008). Given that transparency and local
populations’ participation in the selection of the project sites are essential conditions for the
development of this sector, we propose a negotiation approach based on the coupling of
MultiCriteria Decision Analysis (MCDA) and Geographic Information System (GIS), named
TIMED (Territorial Intelligence Modeling for Energy Development). MCDA establishes a
procedure enabling the participation of various stakeholders involved in the project’s
negotiations, whereas GIS allows the modeling of wind farm implementation scenarios.
Visual information. The human brain treats visual information via vision. The latter is
considered as a data analysis system that facilitates the interaction with the surrounding world
(Milner and Goodale, 1998; Diamant, 2008). Visual cognition allows the identification and
perception of objects and their spatial relations (Pinker, 1985). In fact, visual images
encourage the conceptualization, understanding and appreciation of concepts, thus promoting
17
communication (Yuwei Jiang, 2008) and, in general, any communication of results (Hodgkin,
2005; Manco, 2004). The visualization of information allows a rapid transfer of data and
eases its analysis, thus, facilitating decision making. However, this information treatment
needs to be structured and must objectively represent the decision’s environment and the
decision-makers. More importantly, the rejection or acceptance of this visual information will
greatly affect the decision making (Meyer, 1977).
Wind energy project. The wind energy project is based on technical, financial,
environmental and social aspects. Technical aspects include the identification of regions that
are favourable for the project’s implementation, the activities to be carried out during the
construction process, the capacity of the electrical network, the connection to the latter,
storage, variations in the wind’s direction and strength, among others. Financial aspects
pertain to the economic viability of the project. Environmental aspects include impacts on
fauna and flora and all other factors that may affect the environmental services available to
the area. Social aspects include the use of public and private land, heritage elements,
communication systems and the landscape.
GIS exist since 1960 (Gregory and Ell, 2008). Since then, they have evolved and
grown in importance (Chakhar, 2006). As information technology and the concept of
sustainable development evolved, the GIS became a prime method for territorial management
(Dao, 2005; Malczewski, 2004) as well as for the planning and integration of renewable
energies (Dominguez and Amador, 2007). They rely on spatially referenced numerical
databases with storage, recording, updating, querying and presentation features (Malczewski,
1999; Nobre, 2009; Chakhar and Martel, 2003; Longley, Goodchild, Maguire and Rhind,
21
2011). These databases make the exchange and analysis of information easier (Joliveau
2006). GIS improve the selection of solutions to territorial management problems
(Malczewski, 1999). Territory’s biophysical characteristics must be factored with the socio-
economic and cultural traits of the communities living in this territory (Lovett and Appleton,
2008). This is why territorial management decisions are very complex (Laaribi, Chevalier
and Martel, 1996). However, current GIS do not provide an efficient data analysis when there
are multiple and contradictory criteria. To promote analysis, the GIS therefore needs to rely
on a mechanism that integrates stakeholders’ preferences and proposes choices in the context
of an evaluation of conflicting criteria (Chakar and Martel, 2004); GIS software has begun
offering such mechanisms, resolving this situation. Indeed, GIS-related information
technologies have been developed to better respond to changes in environmental management
and GIS are considered strategic for environmental applications and territorial planning
(Joliveau, 2006; Lovett and Appleton, 2008). We need to emphasize that researchers use
different approaches to coupling MCDA and GIS procedures; for example, IDRISI-GIS and
ESRI Arc GIS, which may use the Analytical Hierarchy Process, the Boolean Overlay or the
Weighted Summation procedure to do this (Lidouh, 2011). Research and development in this
area will continue.
As for MCDA, it is a method based on mathematical tools and certain methods for
comparing alternatives in order to help the decision-making process (Chakhar and Mousseau,
2008a). MCDA was developed mainly after the Second World War. It shifted from cost-
benefit analyses to a multi-objective mathematical programming method and, after, to multi-
criteria analysis (Laaribi, 2000). Local stakeholders participate in the resolution of problems
related to environmental management and territorial planning. For example, this local
participation can take the form of a refusal or acceptance of the project, comprising a
subjective component based on the culture, traditional know-how and specific history of the
local stakeholders. Moreover, MCDA is a useful method to improve decision-making on a
complex social problem (Turcanu, Carlé and Hardeman, 2008). MCDA evaluates both
objective and subjective features by using several criteria. These criteria require the
knowledge of the stakeholders’ preferences (Joliveau, 2006). This way, MCDA integrates
“subjective” preferences, allows the construction of a global preference model and helps
22
identify areas of convergence between different points of view. With the MCDA approach, it
is possible to include or exclude a multi-stakeholder approach; however, an ultimate
objective is to help the decision-makers make their decision (Risse, 2004). In order to achieve
this, a multi-criteria evaluation needs to be set up. This evaluation is based on the use of
importance coefficients and weighting for each criterion, which lead to the construction of
preferences (Roy and Mousseau, 1996; Marinoni, 2006). The types of problems tackled by
the MCDA can be of discrete (Multiattribute Decision Analysis: MADA) or continuous
nature (Multiobjective Decision Analysis : MODA). Discrete problems could have to do with
territorial management. A continuous problem could be the determination of the depth of an
oil deposit. In both MADA and MODA problems, algorithms can be developed for the
selection of the best alternatives (Behzadian et al., 2009). The most popular algorithms for
discrete problems are: weighted summation/Boolean overlay, ideal/reference point,
Analytical Hierarchy Process (AHP), and the Outranking methods (like ELECTRE -
Elimination Et Choix Traduissant la REalité- and PROMETHEE -Preference Ranking
Organisation Method for Enrichment Evaluation-.
each stakeholder. This is the reason why we consider that each given stakeholder can win or
lose in these negotiations (Bassand, 2007). The final decision evolves during the different
stages of negotiation. Different stakeholders can participate at different stages; however, not
all stakeholders become decision-makers.
Between 1991 and 2009, the MCDA-GIS coupling was used in fields such as water
management, territorial management, natural risks, identification of potential environmental
conflicts, forest conservation, urban planning, heritage sites, road infrastructures, waste
management, renewable energy resources, electricity sector, among others. A range of multi-
criteria algorithms already exist within MCDA and the spatial analysis functions of GIS are
well known by researchers. Particularly, the participatory and collaborative multi-decision
makers MCDA-GIS approach has become an unavoidable method for territory management
projects as the participation of local stakeholders is considered to be fundamental for the
success of such projects (Sébastien, 2006). It is necessary to emphasize that the Public
24
Participation GIS (PPGIS) refers to the inclusion of citizens’ knowledge, as well as public
knowledge, to take part in obtaining the results of the application of GIS. Through this
participation it will be obtained a local mapping to be used for different purposes (i.e. policy
development, knowledge production, etc.) (Brown, 2011; Ghose, 2004; Joliveau, 2006;
Sieber, 2006). Collaborative GIS refers to a group organizational techniques to assist in
working together (i.e. electronic voting, scenario design) (Joliveau, 2006).
Among the limitations to the use of MCDA-GIS coupling, we note its technological
requirements. The use of software or participatory websites requires knowledge that
stakeholders may lack in many cultural and educational contexts. This may be the case when
there is a generation gap (teens that are well versed with the internet and virtual social
networks as opposed to the elderly) or a lack of access to information technology (internet,
computers, and software). A final type of limitation concerns the political context in which
the study takes place. Indeed, the first published studies all used a single decision-maker
approach, which was the norm in certain countries where, prior to 1990, social participation
was not yet a common practice. However, nowadays, some countries have yet to include the
participatory aspects in the decision-making process. To apply a participatory and
collaborative MCDA-GIS approach, the political context must be conducive to it. This is why
it is important that countries that already encourage social participation support the
application of this MCDA-GIS approach. This way, it will be possible to gather knowledge
on decision-making process pertaining to territorial management.
the transparency of the analysis results, even though the creation of such databases is a long
and expensive task.
This approach offers other opportunities, including improving the analytic capacities of
GIS; developing the integration of MCDA-GIS software for spatial solutions analysis;
technological development of spatial analysis and geographic information; developing the
design or operational validation of MCDA and/or GIS use; developing or improving a theory
of spatially referenced data; developing multi-criteria aggregation methods; developing
spatial analysis functions, integrating analytical functionalities within the GIS; developing
group dynamics organizational techniques; developing collaborative work support techniques
for the treatment of spatially referenced problems (for example, questionnaires or electronic
voting, which offer different (space/time) options to consultation table participants);
developing or improving spatial analysis theory, developing and/or improving estimate and
decision control techniques added to the GIS as these techniques enable an evaluation and
comparison of GIS-identified sites; enhancing the knowledge of phenomena and procedures
that have an impact on a given territory; using multimedia maps to improve communication
power when presenting analysis results and seeking a consensus solution; acceptance by
planners to work in a collaborative manner with all stakeholders; facilitating access to
information for all stakeholders in a given project and genuine public participation in the
decision-making process and not only reply to questionnaires designed by decision-makers to
implement their own models (Joliveau, 2006).
c) Environmental protection
d) Economic efficiency
f) Access to knowledge
g) Subsidiarity
i) Prevention
j) Precaution
l) Biodiversity preservation
o) Polluter pays
p) Internalisation of costs
sustainability”. This participation and engagement are ensured during the co-development of
the preference systems of the stakeholders involved in the MCDA process.
In section 2.2.2, we have seen that the participation of all of the stakeholders in the
decision-making process guarantees transparency. Section 2.2.3 demonstrates the relevance
of the MCDA-GIS coupling for holding negotiations to seek feasible consensus between
stakeholders, based on their different preference systems. Section 2.2.4 showed that the
application of the MCDA process helps comply with the 16 principles of Quebec’s
sustainable development policy.
This section (2.3) emphasises that several authors have been applying the MCDA-GIS
approach in fields like: energy, radioactive waste, natural hazards, municipal incineration,
watershed management, territory use, vegetation management, natural resource management,
bike path construction, alluvial plain management and others. In particular, among the
numerous works using this approach to analyse problems, we find: Carver, 1991; Martin, St-
Onge and Waaub, 1999; Joerin and Musy, 2000; Chen, Blong and Jacobson, 2001; Merwe
and Lohrentz, 2001; Feick and Brent Hall, 2002; Bussink, 2003; Previl, Thériault and
Rouffignat, 2003; Strager and Rosenberger, 2005; Malczewski, 1999; Malczewski, 2006(b);
Shutidamrong and Lovett, 2008; Simão et al., 2008, and Chang et al., 2009. Just like other
authors before and after, these authors help us identify the factors that lead to transparent and
collective decision-making, offered by the multi-stakeholder and multi-decision-making
MCDA-GIS approach, reaching sustainable development solutions. Among these numerous
works, Banville et al., 1998; Laaribi, 2000; and Joliveau, 2006 focus on MCDA-GIS
coupling factors that all revolve around the following four lines:
MCDA-GIS integration;
Decision validation.
29
MCDA-GIS integration aims at identifying the links that exist between the GIS and the
MCDA software used by the researchers. Support of geographic information processing
allows us to identify database use and map creation factors. The analysis support system
focused on decision-making aims at identifying the tools that the participants used for
decision making. Decision validation aims at identifying whether the decision makers used a
second analysis to verify if the robustness of the first choice is an optimal solution.
Through these four lines, it is possible to examine different aspects of the approach
such as database storage capacity, the participatory construction of the database; the existence
of pre-established, user-friendly or free data; visual aids such as map creation, or virtual 3D
simulations; use of electronic tools to help in collaborative work; the multi-criteria methods
used; model selection or in-depth data exploration tools and techniques, such as sensitivity
analysis; the use of organizational and group dynamics techniques; the creation of knowledge
about procedures that have an impact on territory development; the use of maps as a
communications tool; the acceptance of collaborative work by all of the involved
stakeholders; the measuring of the stakeholders’ influence on the final decision-making
process or the participation of multi-disciplinary experts.
In addition to Joliveau, 2006, Malczewski, 2006(a) identifies the MCDA and GIS
components used by researchers. The GIS components consist of the types of spatial
representation (matrix model, raster, or vector model), the types of spatial criteria (explicit
and implicit), and the spatial alternatives. The MCDA components refer to the types of
decision criteria and algorithms used when applying the MCDA-GIS approach. The MCDA
criteria can be analysed using MADA (multiatribute decision analysis) or MODA
(multiobjective decision analysis). As for the uncertainties, they have an impact on decision
quality (Ascough II et al., 2008). Uncertainties are unavoidable in reality and are related to
the updating of data, the nature of the data objects, distinctions between represented objects
or the context in which they are produced, among others. By mastering these uncertainties,
higher quality data are obtained and this is reflected in the analysis results. In other words,
using all these elements included in the MCDA-GIS multi-criteria and multi-stakeholder
coupling fosters transparency in the decision-making process, helps find consensus between
30
TIMED approach consists of the four following modules: MCDA, participatory and
collaborative GIS, active involvement of the stakeholders and scientific knowledge/local
knowledge. The four modules are represented in a figure containing three circles (see Figure
2.1). The outer circle illustrates the MCDA and GIS procedures. The inner and middle circles
show the stakeholders’ involvement. Taken together, the three circles present the knowledge
of all of the stakeholders. We have chosen to use a circle to illustrate the MCDA-GIS process
as a negotiation facilitator, and to show the continuous and unrestrained feedback during the
stakeholders’ participation and involvement. These stakeholders are placed at the heart of the
process to emphasize their dominant role. The functions carried out exclusively by experts,
such as the choice of the evaluation method and the choice of the evaluation scale, are
represented in a rectangle. Moreover, other functions can be carried out exclusively by
experts, such as criteria selection and identification of indicators, as long as all the
stakeholders consent to this. The arrows tied to a circle indicate the participation of all of the
stakeholders in a given activity such as the construction of a participatory database in the
31
GIS. Arrows crossing circles indicate an activity to be carried out by a specific stakeholder
such as an expert.
All of the stages in the MCDA (see figure 2.2) are indicated in the approach’s outer
circle: problem definition, identification of issues, selection and weighting of criteria,
identification of indicators, evaluation matrices, preference analysis, decision consolidation
and decision.
32
With these stages (see table 2.1), the MCDA will offer answers to the involved
stakeholders’ questions in a real-world decision making process (Roy, 2005). These questions
are about the effects of the different options presented. The representation of preferences in
the real world is undoubtedly one of the fundamental problems of MCDA (Figueira, Grego
and Ehrgott, 2005). This representation is made through the first six stages of the MCDA
presented in the TIMED approach, i.e. problem definition, issues identification and
structuring, selection and weighing of criteria, identification of indicators as well as
evaluation matrices.
33
Another objective of these first stages is to enable stakeholders to identify their own
preferences and concerns in a collaborative manner and to structure them in a coherent family
of criteria. Logically, this multi-stakeholder approach entails a multi-criteria context in order
to assess the environmental, economic and social issues. For example, environmental issues
could be represented, among others, by a pressure on the natural environment criteria; the pay
off of the project could be an economic issue; and social issues could be represented by a
quality-of-life criterion. The criteria serve to structure and model the problem. They need to
be exhaustive, coherent and non redundant. The maximum number of criteria to analyze a
complex problem, as in our case, is around 12 (Roy and Bouyssou, 1993). Once the criteria
are established, they are evaluated according to their qualitative and quantitative nature with
the help of indicators. For example, a “direct loss of habitats on the sites where equipment is
installed” indicator could be designed to measure a “pressure on the natural habitat” criterion.
A “revenue from the sale of produced energy” indicator could measure a “project
profitability” criterion. A “landscape modification” indicator could be created to measure a
“quality of life” criterion. It should be noted that the ranking of the scenarios is based on the
selection of criteria and indicators that take into account the knowledge and values of all of
the stakeholders. The process of defining these indicators and criteria can help resolve
conflicts. Moreover, decision-making support tools imply “the integration of values and
preferences of one or several stakeholders in the decision-making process” (Roy and
Bouyssou, 1993); the preference system of the stakeholders much be clearly expressed and
34
the weighting of the criteria must accurately reflect these values. These preferences are based
on experience, scientific and technical judgement and personal values (Siddayao et al. 1993).
Furthermore, the results of the preference analysis show the arguments for and against
the different wind farm scenarios and these will be ranked by preference. The distribution of
interest groups will also be represented as well as oppositions and coalitions. The
stakeholders may evaluate the effects of criteria weighting on the negotiations. As for the
selection of the preference function, and that of the evaluation scale, they will be performed
by the expert. This is important because the preference function linked to a criterion “allows
the decision maker to amplify or mitigate the impact of this criterion in the decision-making
process” (Brans and Mareschal, 2002). This function will enable the aggregation of
preferences according to an Anglo-Saxon or European concept. The former one is aimed at
“achieving objective truths in decision-making” whereas the latter one seeks “to obtain
structured and coherent results” through a process based on models “which are, at least
partially, co-developed with the decision maker (Roy 2009). We have chosen this latter
option, European, to ensure this co-development with the decision-maker and because it will
help make the decision-making process more transparent. Subsequently, a validation of the
decision is also made. This validation consists in varying the model’s parameter groups
according to different global assumptions (e.g. optimistic and pessimistic assessments). This
is very useful for the decision-maker in order to fully understand the reasons behind the
agreements and disagreements between the stakeholders and to take appropriate support
measures for the implementation of the chosen scenario. Finally, the obtained results will
allow the involved stakeholders to complete the negotiation stage. Depending on expressed
needs, the decision-making support procedure could restart at any of the approach’s stages. It
is the ensemble of the stages, taken together and developed by all of the stakeholders, that
fosters the transparency of the decision-making process of the selection of a wind farm site,
thanks to the sharing of different point of views and the search of shared consensus.
After the selection of the criteria indicators used to assess the issues, all the geo-
referenced variables are identified to be analyzed with the participatory and collaborative GIS
module indicated in the outer circle. For example, locating zones with a good wind potential
35
and areas that are occupied (residential, farming and protected zones, etc.) are important geo-
referenced data to take into account when identifying relevant wind farm implementation
areas. Moreover, the potential of the wind resource is considered an indicator of the “project
profitability” criteria of the economic issue. In general, all the geo-referenced data on the bio-
physical environment and socio-political factors will be taken into account to provide the
most relevant information to the stakeholders and decision-makers. Then, a spatial analysis
should be performed to determine the best potential wind farm sites. To provide these sites,
for example, binary layers (yes/no) could be used and combined by intersection with a results
layer. In addition, the preparation of maps, in collaboration with the stakeholders, will also
provide a visualization of the scenario assessment’s results according to the different
indicators.
This module will therefore include a spatial representation of the scenarios and
physical data, nuance the scenarios’ performance indicators, evaluate the scenarios according
to the MCDA indicators, which are reflected in spatial measurements, and illustrate the
results on maps. The construction of the database will be participatory (experts and
stakeholders), whereas the experts will be in charge of specific tasks like: choosing data
representation (raster or vector), translating raster-to-vector (if necessary), identifying the
appropriate scale, obtaining relevant available data, defining the geometric integration (for
example, the selection of the coordinate system and projection transformation), data
integration (e.g. topology), spatial analysis of data (e.g. score layer production, multi-criteria
analysis) or the cartography.
It is recommended that all of the stakeholders who are affected by the situation
(Baburoglu and Garrr, 1993) be included. We have used four categories to represent the type
of active involvement of all of the possible stakeholders in a territory management project
(Prades, Loulou and Waaub, 1998). These categories are divided into: civil society, private
sector, public sector and experts (Figure 2.4). This division is important due to the fact that
each category has its own preference system. For example, the private sector might prefer a
cost-benefit approach whereas the civil society would insist on the community’s quality of
life. Some stakeholders can belong to more than one category. However, it is precisely the
stakeholder’s support of a category’s preference system, and not that of another preference
system, that determines which category he belongs to. For example, an expert who works for
the private sector (or public sector) could be part of the private sector (public sector) category
because of her/his opinions, whereas an independent expert would probably belong
exclusively to the “Experts” category. However, a simple classification of the stakeholders
remains insufficient when analysing the decision-making process and the relative weight of
the decisional power of each category needs to be determined. This weight could be the same
for each category or unequal (Figures 2.5, 2.6 and 2.7), in which case the most favoured
sector would have more influence on the decision. Even when each category has a different
power of decision weight, the stakeholders’ involvement will foster the analysis of varied
solutions that will enhance the wind farm site selection process. This involvement will also
improve communication and transparency in the decision-making process since clear rules
will be followed from the beginning of the negotiations and during the search for a
consensus, and in this way will ensure a better integration of the wind farm project in the host
community.
37
Figure 2.6 Stakeholder Involvement Module. More Decisional Power for the Public
Sector.
Figure 2.7 Stakeholder Involvement Module. More Decisional Power for the Private
Sector.
39
This module is represented by a figure with three circles (figure 2.8) and its aim is to
build an argument framework on a cognitive basis that takes into account the stakeholders
and their values during the development of wind projects. This cognitive structure must
include both the scientific knowledge of experts and local knowledge. The latter will be
based on objective arguments and will therefore not represent personal opinions. As indicated
above, local knowledge contributes certain particular nuances to the decision-making process
thanks to a specific expertise which, in practice, is related to the local conditions and context.
The outer circle represents the scientific knowledge of experts. This knowledge is related to
specific tasks, such as the selection of the multi-criteria assessment method (algorithm) and
its measure scales or the use of GIS. The experts may also have the task of determining
criteria and choosing the indicators.
The argument framework will be built with the criteria and criteria indicators. These
will be weighted according to the stakeholders’ values (public sector representatives, private
sector representatives, the civil society and experts).
2.6 Conclusion
Certain stages of this approach require that efforts be made, such as building a
participatory database or holding numerous work sessions. However, the MCDA’s theoretical
elements as well as the GIS’ spatial analysis methods are well known by experts and
numerous software are accessible to them, which makes the implementation of the proposed
TIMED approach feasible and adaptable to the management of other renewable energy
resource development projects such as biomass management. Another example of the efforts
to be made is to structure the participation procedure in such way to ensure a serious and
respectful participation among the stakeholders.
The application of the TIMED approach could provide an opportunity for all of the
stakeholders involved in the selection of a wind farm development sites to participate in the
upstream stages of the decision-making process. Once the population is better involved in the
decision-making process the projects’ design will be quicker and, at the same time, more
efficient measures to integrate wind farm projects in host communities will be identified and
established. All these features will encourage the social acceptance of wind farms. This social
acceptance, which has sometimes been compromised, has been the subject of recent research
that has led to requests that development strategies be reoriented and focused on the
community (Jones and Eiser, 2010). Furthermore, to enable the use of the participatory and
collaborative MCDA-GIS approach, the political context must be conducive to it. This is why
41
it is important that governments that already encourage social participation support the
application of this approach. At the same time, it would be possible to gain knowledge on
territorial management decision-making processes.
TIMED approach was created in 2010 and tested in 2011. In this first article, we
present what this approach is about. In a second article, we will present its test and the result
obtained when applied to a wind farm in operation since 2006. This second article appeared,
in 2013, under the title ‘Territorial Intelligence Modeling for Energy Development (TIMED)
– A Case Study for the Baie-des-Sables (Canada) Wind Farm’, in the International Journal of
Multicriteria Decision Making. Moreover, a third article will illustrate how strategic
postulates can be articulated to it, to show a possibility to analyze concerns that can’t be
transformed into criteria. It was published in 2013 as a book chapter, in Assessment and
Simulation Tools for Sustainable Energy Systems, Fausto Cavallaro (Ed.). A fourth work will
be published to confirm the effectiveness of TIMED to create the spaces needed to have
successful community participation. Finally, different works will be made for the transfer and
adaptation of TIMED approach to another socio-cultural-economic and/or environmental
context, another renewable energy sector or another kind of territorial development project.
Acknowledgements
We sincerely acknowledge and thank the Natural Sciences and Engineering Research
Council (NSERC) and the Wind Energy Research Laboratory of the Université du Quebec à
Rimouski (UQAR) for their financial support.
Bibliography
André, P., Delisle, C. and Revéret, J.-P., 2003. L’évaluation des impacts sur
l’environnement : processus, acteurs et pratique pour un développement durable. Presses
internationales Polytechnique. Montréal.
42
Ascough Ii, J. C., Maier, H. R et al., 2008. “Future research challenges for
incorporation of uncertainty in environmental and ecological decision-making”, Ecological
Modelling, 219(3-4): 383-399. Elsevier.
Bailey, D., Goonetilleke, A. and Campbell, D., 2003. “A New Fuzzy Multicriteria
Evaluation Method for Group Site Selection in GIS”. Journal of multi-criteria decision
analysis. 12:337-347. John Wiley & Sons, Ltd.
Banville et al., 1998. “A Stakeholder Approach to MCDA”. Syst. Res. Vol. 15, 15-32.
John Wiley & Sons, Ltd.
Brans, J.-P. and Mareschal, B., 2002. PROMETHÉE-GAIA. Une méthodologie d’aide
à la décision en présence de critères multiples. Éditions de l’Université de Bruxelles.
Brown, G., 2011. ‘An empirical Evaluation of the Spatial Accuracy of Public
Participation GIS (PPGIS) Data’. Applied Geography 34 (2012) 289e294 Elsevier Ltd.
43
Buuren J. and Hendriksen A., 2010. “A Learning and Decision Methodology for
Drainage and Sanitation Improvement in Developing Cities”. Dans Social Perspectives on
the Sanitation Challenge, sous la direction de Vliet, B., Spaargaren, G. et Oosterveer, P.
Dordrecht, Springer Verlag, 2010. New York.
Chakhar, S. and Martel, J.-M., 2004. “Towards a Spatial Decision Support System :
Multi-Criteria Evaluation Functions Inside Geogrpahical Information Systems”. Annales du
LAMSADE No. 2
Chang, N-B. et al. 2009. “Fair fund distribution for a municipal incinerator usign Gis-
based fuzzy analytic hierarchy process”. Journal of Environmental Management 90, 441-454.
Elsevier.
Chen, K., Blong, R. and Jacobson, C., 2001. “MCE-RISK : integrating multicriteria
evaluation and GIS for risk decision-making in natural hazards”. Environmental Modelling &
Software 16, 387-397. Elsevier.
Cinderby, S., Snell, C. and Forrester, J., 2008. “Participatory GIS and its application in
governance: the example of air quality and the implications for noise pollution”. Local
Environment 13 (4): 309-320.
Dao, H., 2005. “Le rôle des systèmes d’information géographique pour le
développement urbain durable”. Dans Enjeux du développement urbain durable.
Transformations urbaines, gestion des ressources et gouvernance, de Da Cunha, A., et al.
Presses Polytechniques et Universitaires Romandes.
Dina, N., Leonid, C. et al., 2009. “Business Systems Modelling: Principles and
Practices”. Integrated Series in Information Systems 19.
Dominguez J. and Amador J., 2007. “Geographical information system applied in the
fiel of renewable energy sources”. Computers & Industrial Engineering 52, 322-326.
Elsevier.
EIA, 2009. International Energy Outlook 2009. Office of Integrated Analysis and
Forescasting. U.S. Department of Energy.
Failing, L., Gregory, R. et al., 2007. “Integrating science and local knowledge in
environmental risk management: A decision-focused approach”. Ecological Economics
64(1): 47-60. Elsevier.
Feick, R. and Brent Hall, G., 2002. “Balancing consensus and conflict with a GIS-
based multi-participant, multi-criteria decision support tool”. GeoJournal 53: 391-406, 2002.
Kluwer Academic Publishers.
Figueira, J., Greco, S. and Ehrgott, M., 2005. Multiple Criteria Decision Analysis :
State of the art Surveys. Springer.
Ghose, R., 2004. ‘SIG et participation citoyenne : des projets de revitalisation urbaine
aux États-Unis’, in S. Roche and C. Caron (Eds.), Aspects organisationnels des SIG.
Lavoisier.
Gomez Delgado, M. and Bosque Sendra, J., 2004. “Sensitivity analysis in Multicriteria
Spatial Decision-Making: A review”. Human and Ecological Risk Assessment, 10:1173-
1187. ASP.
Gonzalez, R. M., 2002. “Joint learning with GIS: multi-actor resource management”.
Agricultural Systems 73(1): 99-111.
Gregory, I. and Ell, P., 2008. Historical GIS: Technologies, Methodologies, and
Scholarship. Cambridge University Press.
Hélimax Énergie Inc. and AWS Truewind LLC 2005. Inventaire du potentiel éolien
exploitable du Québec. Québec. Ministère des Ressources naturelles et de la Faune du
Québec.
Higgs, G., Berry, R. et al., 2008. “Using IT approaches to promote public participation
in renewable energy planning: Prospects and challenges”. Land Use Policy 25(4): 596-607.
Elsevier.
Hodgkin, J., Belton, V. et al., 2005. “Supporting the intelligent MCDA user: A case
study in multi-person multi-criteria decision support”. European Journal of Operational
Research 160(1): 172-189. Elsevier.
Jegen, M., 2008. “L’acceptation sociale des projets éoliens au Québec”. Rapport
mandaté par Ressources naturelles Canada. Gouvernement du Canada.
Joliveau, T., 2006. “Le rôle des systèmes d’information géographique (SIG) dans la
planification territoriale participative”. Dans Aide à la décision pour l’aménagement du
territoire. Méthodes et outils, sous la direction de Graillot, D. et Waaub, J.-P. LAVOISIER.
Joerin, F. and Musy, A., 2000. “Land management with GIS and multicriteria
analysis”. International transactions in operational research 7, 67-78. Pergamon/Elsevier.
IPCC, Pachuri, R. K. and Reisinger, A. (Eds.), 2007a. Synthesis Report. IPCC, Geneva,
Switzerland.
47
IPCC. Metz, B., Davidson, O.R., Bosch, P.R., Dave, R. and Meyer, L. A. (Eds.),
2007b. Mitigation of Climate Change. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA.
Kain, J.-H. and Söderberg, H., 2008. “Management of complex knowledge in planning
for sustainable development: The use of multi-criteria decision aids”. Environmental Impact
Assessment Review 28(1): 7-21. Elsevier.
Laaribi, A., Chevallier, J. J. and Martel, J. M., 1996. “A spatial decision aid : a
multicriterion evaluation approach”. Comput., Environ. And Urban Systems, Vol. 20, No. 6,
pp. 351-366. Elsevier.
Laaribi, A., Chevallier, J. J. and Martel, J. M., 1997. “A Spatial Decision Aid : A
Multicriterion Evaluation Approach”. Comput., environ. And Urban Systems, Vol. 20, No. 6.
Elsevier Science Ltd.
Laaribi, A., 2000. SIG et analyse multicritère. Hermès Science Publications, Paris.
Lidouh, K., 2011. “An Overview of Existing Multi-Criteria Spatial Decision Support
Systems”. European Working Group, Multiple Criteria Decision Aiding, series 3, no. 23.
Lovett, A. and Appleton, K., 2008. Gis for Environmental Decision-Making. CRC
Press. Taylor & Francis Group.
Longley, P.A., 2006. “Grand challenges, environment and urban systems”. Computers,
Environment and Urban Systems 30, 1-9. Elsevier.
Longley, P.A., Goodchild, M., Maguire, D.J., and Rhind, D., 2011. Geographic
Information Systems and Science. John Wiley&Sons, Inc.
48
Malczewski, J., 1999. GIS and multicriteria decision analysis. John Wiley & Son, Inc.
Malczewski, J., 2006b. “Ordered weighted averaging with fuzzy quantifiers: GIS-
based multicriteria evaluation for land-use suitability analysis”. International Journal of
Applied Earth Observation and Geoinformation 8, 270-277. Elsevier.
Manco, G., Pizzuti, C. et al., 2004. “Eureka!: an interactive and visual knowledge
discovery tool”. Journal of Visual Languages & Computing 15(1): 1-35. Elsevier.
Martin, N. J., St-Onge, B. and Waaub, J.-P., 1999. “An integrated decision aid system
for the development of Saint Charles River alluvial plain, Quebec, Canada”. Int. J.
Environment and Pollution, Vol. 12, Nos 2/3. Inderscience Entreprises Ltd.
Merwe, J. H. and Lohrentz, G., 2001. “Demarcating Coastal Vegetation Buffers with
Multicriteria Evaluation and GIS at Saldanha Bay, South Africa”. Ambio, Vol. 30. No. 2, pp.
89-95. Allen Press on behalf of Royal Academy of Sciences /JSTOR.
Milner, D. A. And Goodale, M. A., 1998 The visual brain in action. Edit Oxford
Univ. Press.
49
Minang, P. and McCall, M., 2006. “Participatory GIS and local knowledge
enhancement for community carbon forestry planning: an example from Cameroon”.
Participatory Learning and Action 54(1): 85-91.
Malczewski, J., 1999. GIS and multicriteria decision analysis. John Wiley & Son, Inc.
Nobre, A. et al., 2009. “Geo-spatial multi-criteria analysis for wave energy conversion
system deployment”. Renewable Energy 34: 97-111. Elsevier.
Pidd, M., 1999. “Just Modeling through: A Rough Guide to Modeling”. Interfaces
29(2): 118-132. INFORMS.
50
Prades, J. S., Loulou, R. and Waaub, J.-P., 1998. Stratégies de gestion des gaz à effet
de serre. Le cas des transports urbains. Presses de l’Université du Québec.
Previl, C., Thériault, M. and Rouffignat, J., 2003. “Analyse multicritère et SIG pour
faciliter la concertation en aménagement du territoire : vers une amélioration du processus
décisionnel ?” Cahiers de Géographie du Québec. Volume 47, no. 130, pages 35-61.
Rouse, L. J., Bergeron, S. J. and Harris, T.M., 2007. Participating in the Geospatial
Web: Collaborative Mapping, Social Networks and Participatory GIS. Advanced Information
and Knowledge Processing. Springer London.
Roy, B., 2005. “Paradigms and Challenges”. Dans Multiple Criteria Decision
Analysis : State of the art Surveys, sous la direction de Figueira, J., Greco, S. et Ehrgott, M.
Springer.
Roy, B., 2009. “Deux conceptions de l’aide à la décision”. European Working Group,
Multiple Criteria Decision Aiding, series 3, no. 19.
Roy, B. and Bouyssou, D., 1993. Aide Multicritère à la Décision : Méthodes et Cas.
Economica, Paris.
51
Roy, B. and Mousseau, V., 1996. “A Theoretical Framework for analysisng the Notion
of Relative Importance of Criteria”. Journal of Multi-Criteria Decision Analysis, Vol. 5, 145-
159.
Simão, A. et al., 2008. “Web-based GIS for collaborative planning and public
participation : An application to the strategic planning of wind farm sites”. Journal of
environmental Management, doi:10.1016/jjenvman.2007.08.032. Elsevier.
Steven Pinker., 1985. Visual Cognition. Cambridge, Mass. MIT Press. Reprinted from
Cognition international journal of cognitive psychology, volume 18.
Strager M. and Rosenberger, R., 2005. “Incorporating stakeholder preferences for land
conservation: Weights and measures in spatial MCA”. Ecological Economics 57, 627-639.
Elsevier.
Turcanu, C., Carlé, B. and Hardeman, F., 2008. “On the Constructive Role of Multi-
Criteria Analysis in Nuclear Emergency Management”. Intelligent Decision and Policy
Making Support Systems. Volume 117. Pages 297-313. Springer.
Van den Hove, S., 2006. “Between consensus and compromise: acknowledging the
negotiation dimension in participatory approaches”. Land Use Policy 23(1): 10-17. Elsevier.
Williams, C. and Dunn, C. E., 2003. “GIS in Participatory Research: Assessing the
Impact of Landmines on Communities in North-west Cambodia”. Transactions in GIS 7(3):
393-410.
WWEA, 2009. World Wind Energy Report 2008. World Wind Energy Association,
Germany.
Wolsink, M., 2005. “Wind power implementation: The nature of public attitudes:
Equity and fairness instead of “backyard motives”. Renewable and Sustainable Energy
Reviews 11(6): 1188-1207. Elsevier.
Résumé : Cet article présente l’application d’une nouvelle approche, nommée Développement énergétique
par modélisation et intelligence territoriale (DEMIT), développée en réponse aux préoccupations exprimées par
une partie de la population au sujet des projets éoliens. La prise de décision transparente et participative incluant
tous les acteurs concernés par le développement d’un projet éolien sont à la base de DEMIT. Pour arriver à une
telle décision, deux méthodes multicritères sont nécessaires : l’Aide à la décision multicritère (ADMC) et les
Systèmes d’information géographiques (SIG) participatifs et collaboratifs. Cette méthodologie a été testée pour le
parc éolien Baie-des-Sables, au Québec, Canada, dans un contexte académique et, donc, ne comportant aucun
changement dans l’opération actuelle du parc éolien. Celui-ci a commencé son opération en novembre 2006. Si
l’implémentation de la méthodologie proposée demande des efforts supplémentaires de la part des développeurs
7
M. de L. Vazquez ()
Ph.D student in environmental sciences at the Université du Québec à Rimouski. Wind Energy Research Laboratory.
300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
email: vazquez_rascon.maria_de_lourdes@courrier.uqam.ca
8
Ph.D., Professor, Geography Department, Groupe d’études interdisciplinaires en géographie et environnement
régional Groupe d’études et de recherche en analyse des décisions. Université du Québec à Montréal (UQAM) Case
postale 8888, succursale Centre-ville, Montréal, H3C 3P8, Qc, Canada
email: waaub.jean-philippe@uqam.ca
9
Ph.D. Professor, Mathematics, Computer Science and Engineering Department. Wind Energy Research Laboratory.
Université du Québec à Rimouski (UQAR) 300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
email: adrian_ilinca@UQAR.QC.CA
54
des projets éoliens, elle a, par contre, l’avantage d’impliquer la communauté hôte et de favoriser un démarrage
plus rapide de la construction et de l’opération du parc éolien.
Abstract : This paper presents the application of a new approach, Territorial Intelligence Modeling for
Energy Development (TIMED), developed in response to growing concerns about the social acceptance of wind
farms. Project development is based on procedure transparency and involvement of all stakeholders. To do this,
the TIMED approach involves two multicriteria methods: Multi-Criteria Decision Aid (MCDA) and participatory
and collaborative Geographic Information Systems (GIS). The model was tested in the Baie-des-Sables wind farm
plan located in Quebec, Canada. This was done as an academic research and did not impact the current operation
of the farm, which started back in November 2006. If this approach requires more efforts from the wind farm
developer at the preparation stage it has the advantages of reducing the risks and overall duration of the
infrastructure launching.
3.1 Introduction
In Quebec, between 1997 and 2011, the Office of Public Hearings in Environment
(BAPE – Bureau d’audiences publiques sur l’environnement) identified 15 wind farm plans
that were submitted to public hearings. Since 2010, the conclusions of the BAPE reports are
showing increasing problems of social acceptability that require further studies and the
development of appropriate solutions to be implemented. Specific aspects that should be
considered are the perceived and real psychosocial impacts of wind farms and the
deterioration of relations between supporters and opponents. Voluntary and upstream use of
the proposed TIMED approach provides the ability to analyze alternative scenarios and build
consensus in decision-making that eventually produce an acceptable if not optimal plan for
all stakeholders.
As of January 2012, the total wind power installed in Canada is 5,265 MW, or
approximately 2% of the country's total electricity demand 10, including 918 MW installed in
Quebec. The province’s Energy Strategy 2006-2015 (Department of Natural Resources and
Wildlife of Quebec-MRNF, 2006) points out that the development of wind power is a priority
10
Canadian Wind Farms. [online] Availablet at http://www.canwea.ca/farms/index_e.php (Accessed 24
November 2011).
55
among the different energy resources and sets a target of 4,000 MW, which is limited due to
technical constraints (approximately 10% of total installed power). As a result of this wind
power development strategy, Quebec has nine wind farms in operation and 18 wind farms
planned to be commissioned between December 2011 and 201511.
Among the different wind farms currently in operation, Baie-des-Sables has been
considered a model that shows the importance of adequate communication and collaboration
with the host community (Natural Resources Canada, July 2008). This wind farm consists of
73 wind turbines of 1.5 MW, for a total of 109.5 MW. It covers an area of approximately
5,200 ha on agricultural and forest lands of the municipalities of Baie-des-Sables (83% of the
wind farm) and Métis-sur-Mer (for the remaining 17%)12. These municipalities are located on
the south shore of the St. Lawrence River, along Highway 132, in the Gaspé Peninsula. The
surrounding natural environment is considered exceptional and the renowned beauty of its
landscapes is an important asset of the local tourism industry. In 2005, due to concerns raised
by the implementation of the wind farm, the Quebec Government’s Office of Public Hearings
on the Environment (BAPE) conducted hearings. The inquiry and public hearing report
determined that the wind farm was acceptable with conditions to be met and recommended,
among other things, that the location of some of the wind turbines be optimized (BAPE,
September 2005).
TIMED approach (Figure 3.1) was presented at the 70th Meeting of the European
Working Group on "Multiple Criteria Decision Aiding," in Moncton, Canada, in September
2009 and at the First Intercontinental Conference on Territorial Intelligence, at Gatineau,
Canada, in October 2011. This approach provides decision makers with a set of conditions
for a transparent and participatory process during the implementation of wind farms. TIMED
put in place four modules: MCDA, participatory and collaborative GIS, contributory
stakeholder involvement (CSI) and scientific knowledge / local knowledge (SK-LK).
11
Projets éoliens au Québec. [online] Availablet at http://www.mrn.gouv.qc.ca/energie/eolien (Accessed
24 November 2011).
12
Wind Farms Baie-des-Sables Description. [online] Availablet at
http://www.cartierenergie.com/en/cartier-energie.aspx?sec=7 (Accessed 7 December 2011).
56
Moreover, to be closer to reality, these four modules are articulated with scenario modeling
postulates as well as decisional weighting for every stakeholder involved in the decision-
making process.
The participatory and collaborative GIS module is used to analyze the georeferenced
variables involved to design the scenarios and then to perform their strategic assessment
57
according to the MCDA module. It also allows for the mapping and communicating of the
results. Thus, this module supports the design, selection and representation of alternative
scenarios or solutions to problems involving territorial management (Laaribi, 2000;
Malczewski, 1999). The creation of different layers of data is essential to the analysis of these
variables and this is done using ArcGis software (ESRI).
community. Finally, the fourth, “Taxation,” refers to elements related to income sharing.
Thus, the postulates are framing the application of the TIMED approach. In this paper, we
address the first postulate exclusively: the current situation. The other three postulates will be
discussed in a subsequent work.
This procedure started with an information meeting with all the stakeholders involved
in the Baie-des-Sables wind farm. The objective of this meeting was to explain the whole
approach, its objectives and expected results, to specify the roadmap and agree on the
meeting schedule. The first meeting took place on March 18, 2010 and the last on September
29, 2011. During the second meeting, the MCDA approach was explained to the participants
and illustrated with the software to be used. A total of eight meetings, three hours each, were
held after the second meeting. Six meetings were carried out with “local stakeholders” (i.e.
farmers, citizens) directly in one of the host municipalities of the wind farm. Two other
meetings, with “external stakeholders” (i.e. the project promoter, experts) were held in
Montreal.
This first stage is divided in two parts: definition of the problem by the stakeholders
and definition of the problem by the decision-aiding team. Considering the wind farm has
been in operation since November 2006, and this work was carried out between 2010 and
2011, this exercise gave the stakeholders the opportunity to review their initial concerns and
transpose them to the current situation. At this stage, the participants emphasized concerns
related to local economic and employment issues, the economic impact of the wind farm, the
numerous wind farms implemented in neighboring regions in which tourism is an important
economic activity, the ownership of the wind resource and the type of ownership of the wind
farms, the common good and the type of wind development as well as participation of the
local stakeholders in the decision-making process, among others. On the other side, the
decision-aiding team defined ways to address the problem: analysis of briefs, invitation to the
stakeholders, schedule of activities and others.
60
The decision-aiding team identified the concerns expressed in the 33 briefs of the 2005
BAPE public hearings, and at the same time took into account the positions of the promoter.
Once these concerns identified, they were presented to the negotiation table to initiate the
discussions. Among others, the issues discussed included employment, wellbeing, costs,
profitability of the wind farm, preservation of the environment, waste treatment, sustainable
development, technology used, safety and health.
The decision-aiding team sorted the issues into one of three categories: sociocultural,
economic and environmental (see Table 3.1). We identified a total of 137 concerns, including
86 sociocultural, 34 economic and 17 environmental. Consequently, it can be observed that
the emphasis is on sociocultural issues. Regarding the economic issues, it has been mentioned
that the host community had less benefits when compared to the larger surrounding region.
As for the environmental impacts of the wind farm, even though this category included only
17 of the 137 concerns, these were always considered negative at the local level.
61
requires from the involved stakeholders only “particularly simple and clear” information
(Brans and Mareschal, 2002) and it has a great visual quality when the results are presented.
The selection of the criteria (Table 3.1) stems from how the concerns are structured
(see appendix C Construction des critères). A criterion is defined as a judgment factor on the
basis of which one measures and evaluates an alternative, and it is related to the stakeholders'
preferences whereas a variable is not necessarily related to their preferences (Laaribi, 2000).
A coherent, comprehensive and non-redundant family of criteria proposed for the negotiation
of the wind turbine installation sites was defined by identifying, in the scientific literature,
answers to the concerns expressed by the stakeholders. Moreover, each criterion has to be
maximized or minimized to indicate that, between two alternatives, we prefer the one that has
the highest (or lowest) score on a scale associated with the criterion (Oberti, 2004). Issues
that cannot be expressed as criteria correspond to the abovementioned postulates.
The set of criteria is a simplified way to represent the vast array of issues. An indicator
helps measure the criteria in a qualitative or quantitative way (Table 3.2). The design of the
indicators was based on a documentation review that included, among others, scientific
reports about the effect of wind turbines on human health or flora and fauna, and about the
63
social acceptability of wind farms in Canada, conclusions of BAPE reports published over
the last 14 years of public consultations on wind farms in Quebec, and documents on the
characterization of the landscape of Quebec.
Indicators ScPa03, ScPa04 and ScPa05 (Fortin and LeFloch 2010; Fortin, Devanne and LeFloch, 2010):
**** Collaboration: Marie José Fortin. Research Chair of Canada in Regional and Territorial Development,
UQAR
A preference scale “is an entirely ordered set of levels associated with a criterion that is
used to assess the effect that actions have on this criterion” (Oberti, 2004). These scales were
designed based on a documentation review and available data (Table 3.2). Most of the
constructed scales are quantitative and the qualitative scales were transposed into numerical
values.
64
The Evaluation Matrix (Table 3.3) is a two-dimensional table in which criteria are
rated according to their performance in each proposed scenario (see section 3.2.2.3 Design of
the scenarios). The ratings correspond to the preference scale. This performance table
represents “an irreplaceable basis for dialogue between stakeholders” (Roy and Bouyssou,
1993).
CRITERIA
ScSS01 ScSS02 ScPa03 ScPa04 ScPa05 ScVa06 ScNA07 EcCo08 EcCo09 EcCo10 EcRe11 EnPh12 EnBi13
Scenario 1 158 150 2 2 2 1 159 139,5 329 160 98,4 80 460
Scenario 4 0 0 4 4 4 4 0 0 0 0 0 0 0
The decision-aiding team is responsible for the D-Sight software runs that integrate the
various parameters (i.e. criterion and stakeholder weights, thresholds) to aggregate them and
thus obtain rankings of the proposed scenarios. At the negotiation table, participants examine,
discuss and negotiate about those rankings, the effects of criteria weighting, patterns
indicating conflicting criteria or stakeholders’ coalitions, and converging or diverging points
of view.
Sensitivity Analysis
This analysis consists in “repeating the original multi-criteria analysis by varying the
values originally assigned to the method’s different parameters” (Maystre, Pictet and Simos,
1994). In this way, it becomes possible to observe “which minor modifications change the
proposed solution” (Maystre, Pictet and Simos, 1994). PROMETHEE computes interesting
results related to weight sensitivities which are made available by two D-Sight tools for user
friendly analysis. A criteria weighting stability interval table provides the range above
65
(under) which the weight of each criterion is challenging the ranking for a given number of
positions (Brans and Mareschal, 2002). This is very useful to plan which sensitivity analysis
has to be performed. Then, a walking weight tool helps visualize new rankings according to
the progressive scaling of the criterion’s weight.
Robustness Analysis
This analysis seeks to “determine the range of parameter variations within which a
recommendation is stable” and it is used to “develop recommendations that are as synthetic
as possible, and acceptable to a wide range of parameter values” (Maystre, Pictet and Simos,
1994), thus separating the strong conclusions from the weak conclusions. The parameter to be
tested in this case study is the weighting of the criteria, although others parameters (i.e.
criteria thresholds), could also be tested.
3.2.1.12 Decision
A co-construction process offers the promoter the opportunity to communicate with the
host community and to build the plan in a shared vision while favoring its improvement using
the greater knowledge that local stakeholders have of their territory. At the planning stage,
the TIMED approach provides the promoter with a rigorous process, based on well-known
procedures used worldwide for decades to communicate with the stakeholders. During the
assessment, TIMED allows the involved parties to build a justification framework to better
guide the discussion on the plan’s implementation, observe how local stakeholders perceive
it, rapidly and synthetically recognize conflicting issues, create different possible scenarios,
identify opportunities to negotiate criteria, and determine the most robust solutions. This
enhances the information that the promoter already has, opens new vistas and guides the
decision-making authorities regarding whether or not to allocate the permit to implement the
plan. However, the final decision to go ahead or not always belongs to the promoter.
This module is participative owing to the fact that the geospatial indicators (thematic
layers) were presented to the negotiation table, in order to allow a group discussion on the
elements of the territory that are to be analyzed. This helps build the collective confidence in
66
the results (Joliveau, 2006). Techniques supporting collaborative work were implemented,
such as the use of emails (stakeholders in different places at different times). However, the
regular postal service was used to communicate with two of the stakeholders.
The following layers were identified as necessary to the GIS analysis: roads to be built,
underground and overhead electrical lines to be built, wind resources, hydrology, localization
of Road 132, paths and snowmobile trails, cross-country ski trails, camping areas, homes,
diversity of birds, sensitive environments, agricultural areas in the wind farm, sugar bushes
and any other type of wooded area as well as the classification of the landscapes of Métis-
sur-Mer and Baie-des-Sables inside the wind farm area.
Data availability, time needed and human resource involvement might be critical
variables and should not be underestimated. The burden of database development and
management is usually imposed on the promoter who has the responsibility to conduct the
environmental assessment studies and obtain authorizations and permits. In this case study,
the researcher in charge had to personally bear the cost of developing the database.
The database includes a digital topographic map of the study area and 12 layers
resulting from the digitalization of PDF documents, mostly from the public domain,
converted to the TIF format. This database was validated by the stakeholders participating at
the negotiation table and it can allow for the preservation, updating, questioning, and
mapping of the data (Malczewski, 1999) and the facilitation of exchanges and information
analysis.
The decision-aiding team handles the D-Sight software, while the stakeholders
determine the design of the scenarios. These are designed by taking into account the
proximity to the residences, sugar bushes and roads together with new electrical lines to be
constructed in agricultural zones, the speed of the wind and the type of land that is affected.
Exclusion zones (buffers) were created to ascertain the potential wind turbine installation
67
sites. The distances were determined by taking into account a documentation review (i.e.
regulations of the municipalities of the Gaspésie). Scenario design requires: data loading and
integration, pre-processing (i.e. buffer creation or weighting), processing (operations on the
tables and layers), and production of output. The results showed that if a separating distance
of 1 km between the wind turbines and critical territorial components had been fixed, it
would have almost completely prevented the implementation of the wind farm. After
discussions, the negotiation table retained four scenarios:
1. Scenario 1: the wind farm as it was built (73 1.5-MW wind turbines);
2. Scenario 2: elimination of the wind turbines that are located within 500 meters
of homes or 1000 meters of route 132, and, in addition, taking into account the
municipal regulations regarding distance constraints (38 1.5-MW wind
turbines);
The maps (see example on Figure 3.2) that were created helped make the stakeholders
aware of any relevant information. To enrich the meeting discussions, with the ArcGIS
software, we worked directly on layers to visualize and highlight elements that were
originally not included in the maps. This helped improve communications as well as the
analysis of ideas that had been expressed verbally. For example, to help design the scenarios,
different hypotheses regarding the separation distances of the wind turbines were tested. This
information was used to design different scenarios with other separation distances and
different numbers of wind turbines. Other maps were also created.
68
The specific steps in which the different stakeholders were involved in the construction
of the TIMED procedure are identified in Table 3.4. This table also shows the steps during
which the decision-aiding team was intervening even though this team is not considered a
stakeholder.
Scientific knowledge and local knowledge interact in each stage of the process, support
the multicriteria analysis, generate new knowledge and sustain the software technical
framework. For example, local know-how helped answer questions about the productivity of
the sugar bushes, the significant places according to the local population, the importance of
having a grade school in the region, effects on tourism or on the relationships between
citizens before and after the arrival of the wind farm. Moreover, it also helped plan the work
agenda with the stakeholders: during harvest, it is impossible to hold meetings during the day
because farmers are too busy. Scientific knowledge also allowed the stakeholders to learn,
among other things, about new research findings on bird migration, the newest wind energy
69
technology and estimates of the greenhouse gas emissions produced over the life cycle of a
wind turbine.
Decision-
No. Module Steps Promoter Stakeholders
aiding team
1 MCDA Problem definition (experts)
Problem definition (negotiation
table)
2 MCDA Identification of issues
3 MCDA Structuring of issues
4 MCDA Selection of evaluation method
5 MCDA Selection of criteria
Selection of criteria (validation)
6 MCDA Weighting of criteria
7 MCDA Identification of indicators
Validation of indicators
8 MCDA Selection of measurement scales
9 MCDA Evaluation matrix
10 MCDA Preference analysis
11 MCDA Consolidation of judgment
12 MCDA Decision
13 GIS Selection of georeferenced
information
14 GIS Development of database
Validation of database
15 GIS Construction of scenarios
16 GIS Preparation of maps
17 CSI Selection of decision-making
power
18 SK-LK Sharing
3.3 Results
In this paper, we present the rankings obtained by three models. Each model is based
on the same postulate (existing framework) (see section 3.1.3), the same four scenarios (see
section 3.2.2.3), the same evaluation matrix (see section 5.2.1.9), and three different
weightings of decision-making power (see section 3.1.2). Because of this, three models were
configured with the D-Sight software.
Figure 3.3 (interpreted from View Global Results) shows the results of the global
ranking of Model 1, where each stakeholder has received the same power balance, regardless
70
of the category to which they belong. In this figure, scenarios 3 (38 3-MW turbines) and 4
(do nothing) are also preferred and would be subject to negotiations. The objectives of these
negotiations would be to reduce the number of turbines to be implemented and to examine
possible existing technological alternatives in order to -maintain the planned productivity of
the wind farm. Scenarios 1 (73 1.5-MW turbines) and 2 (38 1.5-MW turbines) are simply not
favored because of their low productivity and the distribution of the wind turbines.
analysis of economic criteria in GAIA shows that jobs and local financial benefits are an
important criteria for the host community.
The D-Sight software weight interval stability table shows that, for seven of the 11
stakeholders, the socio-cultural and environmental criteria have a large interval; this means
that they are not sensitive to weight variation. However, the economics criteria have a smaller
interval and are therefore good candidates to perform sensitivity analysis. To illustrate how
this analysis was carried out, we ascertained that, in the case of one of the civil society
stakeholders, if the total weighting of the economic criteria is increased from 30% to 40%,
the ranking changes and the preferred scenario becomes scenario 3 (construction) instead of 4
(do nothing). This weighting change allows for a redistribution of the relative weight of the
different economic criteria and thus, negotiations to increase by 8.3% the number of local
jobs. In addition, it would also be possible to negotiate to increase by 5% the local financial
benefits criteria. This would improve the local economic benefits compared with the regional
economic benefits.
72
We repeated the sensitivity analysis several times for other stakeholders regarding the
same economic criteria. We observed that, for two stakeholders, the shift from scenario 4 to
scenario 3 requires only to increase by 1.4% and 2.4%, respectively, the overall weight of the
economic criteria. For the rest of the stakeholders, this transition would require that the
overall weight of the economic criteria be increased by 8.4% and even 19.3%. In fact, the
indifferent ranking of scenario 3 and 4 in Model 1 can be considered robust.
Regarding the results of Model 2, each of the four stakeholder categories was given a
decision-making power weight equal to 25%, regardless of the number of stakeholders
included in each of the categories. The global ranking obtained was then scenario 3, followed
by scenario 4, scenario 2, and scenario 1.
Regarding the results of Model 3, the weights were divided as follows: promoter, 40%;
landowner, 10%; public sector, 25%; experts, 12.5%; and civil society, 12.5%. The ranking
obtained was scenario 3, followed by scenario 2, scenario 4 (2 and 4 near equal), and scenario
1.
73
3.4 Discussion
Moreover, in the TIMED approach, to better reflect the stakeholders’ preferences, final
decision-making also benefits from the application of postulates (see section 3.1.3). These
require the creation of 21 others models that will be made in a subsequent work and will also
enrich the analysis of this case study. A SWOT analysis (Strengths, Weaknesses,
Opportunities, and Threats) of the TIMED approach is forthcoming which will suggest
improvements for future use.
3.5 Conclusion
First of all, this test has revealed that there are fixed parameters in the wind farm’s call
for tenders. Therefore, it is useless to consider any scenario that includes a technological
solution that is different from the one already established (i.e. a transition from a 1.5-MW
turbine to a 3-MW turbine). This shows that there are technical and economic limits to what
can be negotiated. However, this needs to change since various stakeholders increasingly
recommend that the participation of the community be promoted in order to minimize
negative impacts and maximize the benefits of wind energy.
74
Acknowledgements
We sincerely acknowledge and thank the Natural Sciences and Engineering Research
Council (NSERC) and the Wind Energy Research Laboratory of the Université du Quebec à
Rimouski (UQAR) for their financial support.
75
References
André, P., Delisle, C. E. and Revéret, J.-P. (2010) L’évaluation des impacts sur
l’environnement. Processus, acteurs et pratique pour un développement durable. Presses
internationales Polytechnique, Canada.
Figueira, J., Greco, S. and Ehrgott, M. (2005) Introduction Multiple Criteria Decision
Analysis : State of the art Surveys, Springer Science + Business Media, Inc., United States of
America.
Fortin, M-J. and LeFloch, S. (2010) «Contester les projets éoliens au nom du paysage :
Au-delà du nimby, discuter du développement des territoires», Globe : Revue internationale
d’études québécoises, numéro Énergie et citoyenneté. Vol. 13 (2) : 27-50.
(http://id.erudit.org/iderudit/1001129ar).
Fortin, M-J., Devanne, A-S. and LeFloch, S. (2010) «Le paysage politique pour
territorialiser l’action publique et les projets de développement : le cas de l’éolien au
Québec», revue Développement durable et territoire [En ligne], Vol. 1, n° 2 | Septembre
2010. (http://developpementdurable.revues.org/index8540.html).
GIRARDOT J.-J. (2009) ‘Results and prospects of caENTI the coordination action of
the Network of Territorial Intelligence’ in CaENTI final seminar in Brussels, Brussels.
76
Joliveau, T. (2006) ‘Le rôle des systèmes d’information géographique (SIG) dans la
planification territoriale participative’, in D. Graillot and J.-P. Waaub (Eds.), Aide à la
décision pour l’aménagement du territoire. Méthodes et outils, LAVOISIER, Paris.
Malczewski, J. (1999) GIS and multicriteria decision analysis, John Wiley & Son,
Inc., United States of America.
Ministry of Natural Resources and Fauna of Quebec, MRNF. (2005) Guide pour la
réalisation d’une étude d’intégration et d’harmonisation paysagères. Projet d’implantation
de parc éolien sur le territoire public. [online] Direction générale de la gestion du territoire
public. Gouvernement du Québec. Available at:
http://www.mrnf.gouv.qc.ca/publications/territoire/programme/projet-eolien.pdf (Accessed
November 24, 2011).
Prades, J. S., Loulou, R. and Waaub, J.-P. (1998) Stratégies de gestion des gaz à effet
de serre. Le cas des transports urbains, Presses de l’Université du Québec, Sainte-Foy.
Natural Resources Canada. (July 2008) L’acceptation sociale des projets éoliens au
Québec. [online] Report by Maya Jegen. Gouvernement du Canada. Available at:
http://www.politique.uqam.ca/upload/files/PDF/Rapport_mandate.pdf (Accessed 25
November 2010).
Roy, B. and Mousseau, V. (1996) ‘A Theoretical Framework for Analysing the Notion
of Relative Importance of Criteria.’ Journal of Multi-Criteria Decision Analysis, Vol. 5, 145-
159. [online]. Available at: http://isi.lamsade.dauphine.fr/scripts/FILES/publi19.pdf
(Accessed 4 March 2010).
CHAPITRE IV
Published in Assessment and Simulation Tools for Sustainable Energy Systems, Fausto
Cavallaro (ed.), 2013, SPRINGER
Résumé: Le développement durable des projets éoliens inclut la prise en compte de la variable socio-
culturelle. Celle-ci peut être décelée à partir des préoccupations citoyennes par rapport à l’exploitation de cette
ressource énergétique renouvelable. Dans cet article, de telles préoccupations ont été incluses dans un processus
d’aide à la décision multicritère et, exprimées comme postulats, elles sont le sujet d’une analyse de robustesse.
Nous décrivons ici une méthodologie pour ce faire et nous l’appliquons comme un cas d’étude au parc éolien de
Baie-de-Sables au Québec, Canada. Il s’agit d’un test académique qui n’affecte pas l’opération du parc installé
depuis novembre 2006. Cependant, l’analyse présentée permet de conclure que, si certaines préoccupations
avaient été prises en compte, un autre scénario de parc éolien aurait eu une note plus élevée. L’analyse de
robustesse des postulats outils de communication et type de propriété de parc éolien a permis d’identifier des
règles objectives sur le classement des différents scénarios. Le nouveau classement a été vérifié à l’aide de
matrices d’évaluation utilisant des valeurs différentes, maximales et proportionnelles par rapport à la matrice
13
M. de L. Vazquez ()
Ph.D student in environmental sciences at the Université du Québec à Rimouski. Wind Energy Research Laboratory.
300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
email: vazquez_rascon.maria_de_lourdes@courrier.uqam.ca
14
Ph.D., Professor, Geography Department, Groupe d’études interdisciplinaires en géographie et environnement
régional Groupe d’études et de recherche en analyse des décisions. Université du Québec à Montréal (UQAM) Case
postale 8888, succursale Centre-ville, Montréal, H3C 3P8, Qc, Canada
email: waaub.jean-philippe@uqam.ca
15
Ph.D. Professor, Mathematics, Computer Science and Engineering Department. Wind Energy Research Laboratory.
Université du Québec à Rimouski (UQAR) 300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
email: adrian_ilinca@UQAR.QC.CA
79
originale. L’analyse de robustesse sur ces matrices d’évaluation a permis d’identifier, dans une situation
conflictuelle, des opportunités pour débloquer des obstacles lors de l’implémentation d’un parc éolien.
Abstract : Sustainable wind energy development takes into account socio-cultural variables that can be
identified from citizens’ concerns about the use of this renewable energy. These concerns are included in a
multicriteria decision aid process, and, expressed as postulates in this study; they are subject to a robustness
analysis. The approach is described and applied to the Baie-des-Sables (Canada) wind farm case study. While this
academic post-installation assessment does not affect the current operation of the farm which started back in
November 2006, we conclude that if these concerns were considered, another wind farm scenario would have got
a higher rating. Robustness analysis with respect to communication tools or type of ownership of the wind farm
made it possible to identify objective rules based on changes in the ranking of scenarios. This change was verified
using evaluation matrices containing different, maximum and proportional values with respect to the values of the
original matrix. The robustness analysis results made it possible to identify, in a conflict situation, opportunities to
remove obstacles to wind farm implementation.
4.1 Introduction
for robustness analysis are described. Additional results for the case of Baie-des-Sables wind
farm are computed and discussed to illustrate actual situation.
MCDA is a procedure that helps solve a complex decision problem. The main stages of
this procedure are: recognition of the stakeholders, definition and development of scenarios,
definition of criteria and their relative importance, preparation of evaluation matrices,
application of a comparison procedure, establishment of the final ranking, robustness and
sensitivity analysis (André et al. 2003; Roy and Bouyssou 1993). The MCDA improves
decision making and eases negotiations by incorporating non-economic objectives and
focusing on well-defined and well-targeted objectives (Corsair et al. 2009). When used in a
participatory manner that involves all of the stakeholders in the decision making process, it
fosters a learning process based on local know-how and systematic knowledge as a means to
reach sustainable solutions to the analyzed problem (Buuren and Hendriksen 2010). This is
therefore an MCDA process in a multi-stakeholder and multi-decision maker context. It is
applied to territorial management and in any situation involving resource-use conflicts, and
this, in a participatory and collaborative perspective. In other words, whereas the single
decision-maker approach tends to be hierarchically vertical and techno-scientific, the multiple
decision-maker approach integrates different knowledge systems (scientific knowledge and
knowledge of local stakeholders).
The analysis relies on spatially referenced numerical databases with storage, recording,
updating, querying and presentation features (Malczewski 1999; Nobre 2009; Chakhar and
Martel 2003; Longley et al 2011). These databases make the exchange and analysis of
information easier (Joliveau 2006). GIS improves the selection of solutions to territorial
management problems (Malczewski 1999). Territory’s biophysical characteristics must be
factored with the socio-economic and cultural traits of the communities living in this territory
(Lovett and Appleton, 2008). This is why territorial management decisions are very complex
81
(Laaribi et al. 1996). However, current GIS do not provide an efficient data analysis when
there are multiple and contradictory criteria. To promote analysis, the GIS therefore needs to
rely on a mechanism that integrates stakeholders’ preferences and proposes choices in the
context of an evaluation of conflicting criteria (Chakar and Martel 2004).
Robustness analysis, which is carried out before the final decision is made, is one of
the main steps of MCDA, and aims at establishing the right basis for a recommendation. Its
main objective is to verify whether the recommendation resulting from the multicriteria
procedure is robust, and if not, to identify what changes produce this result (Roy and
Bouyssou 1993). This analysis seeks to “determine the range of parameter variations within
which a recommendation is stable” and it is used to “develop recommendations that are as
synthetic as possible, and acceptable to a wide range of parameter values” (Maystre et al.
1994), thus separating the strong conclusions from the weak ones. The parameters to be
tested can be the weighting of the criteria or others.
Indeed, it is recognized that the concept of robustness is vast and that there are
different approaches, including requiring that the solution satisfy the robustness conditions
for it to be considered robust (Aloulou et al. 2005). In addition, it is also recognized that
robustness depends on the problem’s context (Vincke 1999).
83
One must also consider that the term “robustness concern” is broader than “robustness
analysis” because this concern is already present “in the problem formulation phase” and it
aims at identifying weaknesses in the formal problem representation (Roy 2007), to
determine whether the recommendation is or not solid.
The authors have established (Vazquez, Waaub and Ilinca, 2013a) that the stakeholders
expressed two kinds of concerns: those that can be translated into evaluation criteria of the
selected turbine sites and those related to the strategic development of the wind power sector.
The latter can become socially controversial issues at a specific wind farm negotiation table,
and end up being part of its specific analysis (Côté 2011) during the discussions on the
turbines’ site selection.
The above-mentioned paper (Vazquez, Waaub and Ilinca,. 2013a) includes the results
produced by the application of the MCDA to the first kind of concerns that are also included
in the current framework used for the implementation of wind farm in Quebec.
The purpose of this contribution is to answer the following questions: What happens
with the strategic concerns? How strong is the result of the existing framework regarding
these strategic concerns? Thus, we consider that these questions basically have as much to do
with the formulation of the real problem expressed by the involved stakeholders as with the
above-mentioned robustness “concern”.
strategic concerns. To do this, scenarios’ performance using pre-established criteria will vary
according to the new evaluation matrices.
Matrices subject to robustness analysis were built in 2011 during the test of the
TIMED approach.
The TIMED approach (see Fig. 4.1) is made up of four modules: MCDA, participatory
and collaborative GIS, contributory stakeholder involvement (CSI) and scientific
knowledge/local knowledge (SK-LK). The goal of these modules is to guarantee
transparency and participation to promote a sustainable energy development of wind energy.
Since TIMED approach is an instrument to find consensus between different interests, it also
describes the negotiation procedure to be followed, while ensuring that the different
stakeholders’ knowledge is integrated in the process and that the periods of participation and
involvement are included.
The four modules are contained in three circles (see Fig. 4.1). The outer circle
illustrates the MCDA and GIS procedures. The inner and middle circles show the
stakeholders’ involvement. Taken together, the three circles present the knowledge of all of
the stakeholders. We have chosen to use a circle to illustrate the MCDA-GIS process as a
negotiation facilitator, and to show the continuous and unrestrained feedback during the
stakeholders’ participation and involvement.
These stakeholders are placed at the heart of the process to emphasize their dominant
role. The functions carried out exclusively by experts, such as the choice of the evaluation
method and the choice of the evaluation scale, are represented in a rectangle. Moreover, other
functions can be carried out exclusively by experts, such as criteria selection and
identification of indicators, as long as all the stakeholders consent to this. The arrows tied to a
circle indicate the participation of all of the stakeholders in a given activity such as the
85
All of the stages in the MCDA process are indicated in the approach’s outer circle:
problem definition, identification of issues, selection and weighting of criteria, identification
of indicators, evaluation matrices, preference analysis, decision consolidation and decision.
This module creates opportunities for the sustainable management of energy resources since
it helps to co-build the preferences system of the stakeholders actively involved with the
decision-maker when the consequences of the proposed actions are examined. This module
enables stakeholders to identify their own preferences and concerns in a collaborative manner
and to structure them in a coherent family of criteria. The criteria serve to structure and
model the problem. Once the criteria are established, they are evaluated according to their
86
qualitative and quantitative nature with the help of indicators. The process of defining these
indicators and criteria can help resolve conflicts. Moreover, decision-making support tools
imply “the integration of values and preferences of one or several stakeholders in the
decision-making process” (Roy and Bouyssou 1993); the preference system of the
stakeholders much be clearly expressed and the weighting of the criteria must accurately
reflect these values. Furthermore, the results of the preference analysis show the arguments
for and against the different wind farm scenarios and these will be ranked by preference.
Logically, this multi-stakeholder approach entails a multi-criteria context in order to assess
the environmental, economic and social issues.
The GIS module is participative due to the fact that the geospatial indicators (thematic
layers) were presented to the negotiation table, in order to allow a group discussion on the
elements of the territory that are to be analyzed. The scenarios to be analyzed in this module
are taking into account the proximity to the residences, sugar bushes and roads together with
new electrical lines to be constructed in agricultural zones, the speed of the wind and the type
of land that is affected. Exclusion zones (buffers) can be created to ascertain the potential
wind turbine installation sites. The distances can be determined by taking into account a
documentation review (i.e. regulations of the municipalities of the Gaspésie, Quebec,
Canada). Scenario design requires: data loading and integration, pre-processing (i.e. buffer
creation or weighting), processing (operations on the tables and layers), and production of
output (maps). The maps helped make the stakeholders aware of any relevant information
and to work directly on layers to visualize and highlight elements that were originally not
included in the maps. This helped improve communications as well as the analysis of ideas
expressed verbally. For example, to help design the scenarios, different hypotheses regarding
the separation distances of the wind turbines can be tested. This information can be used to
design different scenarios with other separation distances and different numbers of wind
turbines.
The CSI module includes four categories of stakeholders: civil society, public sector,
private sector and experts. These categories represent the type of active involvement of all of
the possible stakeholders in a territory management project (Prades et al. 1998). This division
is important due to the fact that each category has its own preference system. However, a
87
simple classification of the stakeholders remains insufficient when analysing the decision-
making process and the relative weight of the decisional power of each category needs to be
determined. This weight could be the same for each category or unequal, in which case the
most favoured sector would have more influence on the decision. In this way, clear rules will
be followed from the beginning of the negotiations and during the search for a consensus, and
will ensure a better integration of the wind farm project in the host community.
Moreover, the four modules are associated with scenario modeling postulates as well
as decisional weighting for every stakeholder involved in the decision-making process. In
(Vazquez et al. 2013b), we had explored the possibilities of a first postulate (the Current
Situation postulate) and an evaluation matrix for the existing framework of wind farm was
constructed. An example of the original evaluation matrix will be introduced later (see Table
4.3) and it is directly in this matrix that modifications for robustness analysis are done.
First, the strategic concerns will be formulated using postulates. Second, the
methodology for analyzing each strategic level postulate will be established while also
identifying the elements to be considered, the evaluation matrices related to the number of
necessary models, the criteria which may be modified and that are to be included in the
analysis, the representativeness of the postulates to be analyzed as well as the design of
matrices with three different types of grades.
The strategic concerns were extracted from the official and public documents
submitted to the Quebec Government’s Office of Public Hearings on the Environment
88
(BAPE), in 2005, for a specific wind farm. These concerns were then structured into four
strategic-level postulates, which are located upstream of the multicriteria decision aid
process:
Current situation: describes the current conditions. It is the set of all the
conditions in which wind farms are currently being developed (i.e. call for
tender conditions, annual voluntary contributions, municipal visibility funds,
regulatory framework of separation distances included in the PCRs -
Provisional control regulations).
Ownership: refers to concerns related to the Type of ownership of the wind
farm.
Communication tools: refers to concerns related to the type of approach used
to communicate with the host community.
Taxation and royalties: refers to elements related to income sharing, for
example, the creation of wind farm revenue redistribution and sharing
formulas.
Thus, the concerns that are not part of the existing conditions lead to three others
postulates (Type of ownership of the wind farm, Communication Tools and Taxation). Two
postulates are subdivided. Type of Ownership of the Wind Farm is subdivided into three sub-
postulates: Public, Large Private Company and Public-Private Partnership (this includes
community wind farm as well as joint wind farms involving municipalities and private
companies). Communication Tools is subdivided into three sub-postulates: Information
Meeting, Consultation and Negotiation Table. The postulates Current Situation and Taxation
and Royalties do not have sub-postulates (see Table 4.1). Therefore, there are a total of four
postulates and eight sub-postulates.
Each strategic level postulate defined earlier needs to be analyzed. The Current
situation postulate has already been analyzed (see Vazquez et al. 2013b). It was analyzed for
each one of the four designed scenarios, using one of the three different decision-making
89
power weightings and one evaluation matrix (Original evaluation matrix). It should be noted
that we could have tested all four designed scenarios for each of the three decision-making
power weightings (see Vazquez et al. 2013b); however, in this chapter we will only illustrate
one example with one of the three weightings.
Table 4.1 Models and matrices required according to the different scenarios,
postulates and power weighting options considered in the robustness analysis
Among other things, we must consider that each sub-postulate (or postulate, as is the
case) has its own evaluation matrix. The Current situation postulate is named NOri (N, grade;
Ori, original) and was designed by the multicriteria decision aid team and validated by all the
stakeholders involved in the case study of the TIMED approach (see Vazquez et al. 2013b).
The analysis of the other three strategic level postulates shall be done by making grade
changes in this NOri matrix. Overall (see Tab. 4.1), to carry out this analysis, 24 models were
configured using the D-Sight Software and taking into account: four scenarios, three
decision-making power weighting options, eight sub-postulates and eight evaluation matrices
(see Table 4.1).
royalties
Once the criteria whose evaluation can be modified are identified, it is necessary to
conduct the robustness analyses. To avoid making the 24 models listed in table 4.1, and in
92
order to draw conclusions from them, the most representative postulates were selected. This
representativeness of all the models was determined based on the variations in the number of
criteria which can be modified: three for Communication tools, nine for Type of ownership of
wind farm and two for Taxation and royalties. Thus, the robustness analysis will be done on
the Communication tools and Type of property of wind farm postulates. As for Taxation and
royalties, whose variability is lower, it will be the subject of a special comment at the end of
discussion results section.
The grading of the evaluation matrices has undergone one of the two procedures:
1) Grading proposed by the decision-aid team and validated by the negotiation table.
2) Grading proposed by the decision-aid team and not validated by the negotiation
table.
The first case corresponds to the matrix of the Current situation postulate, validated by
the stakeholders participating in the negotiation table, because this table was established in
2010-2011 only to evaluate this postulate.
The second case corresponds to the three others postulates (see Table 4.2), which
require the formulation of new matrices. These will be created through the reformulation of
the NOri matrix while at the same time considering the existing issues related to the use of
communication tools, the type of wind farm property and the criteria whose evaluation is
likely to be modified. To make this reformulation, the decision-aid team changed the grades
of these likely criteria. All the other criteria kept their original values validated by the
negotiation table.
93
To start the analysis that will answer whether the results of the Current situation
postulate are robust compared with the other three strategic postulates, it is necessary to
consider three new criteria performance grades:
This produces three new evaluations matrices, which increases further the number of
models to create. Thus, the NDif matrix (see Table 4.4) refers to different grades, the NMax
matrix (see Table 4.5) refers to maximum grades and the NPro matrix refers to proportional
grades. This last one is combined with the NMax matrix to create the NMax-NPro matrix
(see Table 6.6). Each of the three postulates (or sub-postulates) to be analyzed uses the NDif,
NMax and NMax-NPro matrices, respectively.
The utility of having different value performance grades is discussed with the
stakeholders and with the expert MCDA team, as well as values that are reasonable to be
analyzed based on available resources (i.e. funding, time, etc.). The objective is to analyze
different performance scenarios for each postulate and to compare them in order to test the
robustness of a decision. In our test, these value performances were related to a specific
context and should assist stakeholders to understand what happens if we change the
evaluation matrices value by grading the scenario performances in a specific manner. Basic
mathematical concepts (different, maximum, and proportional) were then chosen to easily
illustrate these questions, using whole numbers and percentages. Nevertheless, depending on
the context, another value performance grades could be used if the stakeholders and the
MCDA team agree (i.e. concepts as the average or median) but it is important to have every
time the control over the number of new evaluation matrix to be designed because their
number must be in accordance with the available resources.
not and to determine if a better scenario could be obtained depending of the type of
ownership of wind farm.
4.3 Results
In this matrix (see table 4.3, figure 4.2), all criteria and their grades were validated by
the negotiation table, in different bases. For example, the criterion ScPa03 (significant places)
is rated 4 in scenario 4 because the non-construction of the wind farm allows all of the
activities that took place before the advent of the wind farm to be carried out without
restrictions.
CRITERIA
ScSS01 ScSS02 ScPa03 ScPa04 ScPa05 ScVa06 ScNA07 EcCo08 EcCo09 EcCo10 EcRe11 EnPh12 EnBi13
Scenario 1 158 150 2 2 2 1 159 139,5 329 160 98,4 80 460
Scenario 2 0 78 3 3 3 2 84 87 175 78 51,2 0 220
Scenario 3 0 78 3 3 3 3 108 144 175 106 102,4 0 220
Scenario 4 0 0 4 4 4 4 0 0 0 0 0 0 0
Figure 4.2 Ranking according to the original evaluation matrix of the Current
situation postulate
95
In this matrix, the criteria that are likely to be changed (ScPa03 and ScVa06) received
a new grade. For example, the ScPa03 criterion (significant places) is graded 0 in scenario 4
because no communication tool would be used to discuss about the issues of this criterion.
Similarly, the same criterion is rated 4 in scenarios 2 and 3 since those scenarios are
designed in a participatory manner and using communication tools.
Table 4.4 Evaluation matrix with different grades for the Communication tools
– Negotiation table postulates
CRITERIA
ScSS01 ScSS02 ScPa03 ScPa04 ScPa05 ScVa06 ScNA07 EcCo08 EcCo09 EcCo10 EcRe11 EnPh12 EnBi13
In this matrix, the criteria that are likely to change (ScPa03 et ScVa06) received the
highest possible rating, 4, indicating that in all cases, we maximized the use of
communication tools to discuss issues about values, needs and significant places for people.
CRITERIA
ScSS01 ScSS02 ScPa03 ScPa04 ScPa05 ScVa06 ScNA07 EcCo08 EcCo09 EcCo10 EcRe11 EnPh12 EnBi13
Scenario 1 158 150 4 2 2 4 159 139,5 329 160 98,4 80 460
Scenario 2 0 78 4 3 3 4 84 87 175 78 51,2 0 220
Scenario 3 0 78 4 3 3 4 108 144 175 106 102,4 0 220
Scenario 4 0 0 4 4 4 4 0 0 0 0 0 0 0
Figure 4.4 Ranking of scenarios from Maximum grade evaluation matrix of the
Communication tools – Negotiation postulate
97
The criteria that are likely to change are ScPa03 and ScVa06. In this case, proportional
values are added to each of the chosen criteria in the NOri evaluation matrix. For this
proportional grade, change is also proportional and the ranking obtained is equal to that of the
NOri matrix (therefore there is no need to illustrate the results for this evaluation matrix). In
this case, the PROMETHEE method already gives the answer and there is no longer a need to
make systematic evaluations. This is because, in this case of the ordinary function used for
these criteria, the performance differences between the scenarios remain the same.
To analyze the Type of ownership of wind farm postulate, in addition to the ScPa03 and
ScVa06 criteria, it is also necessary to consider the ScPa05, EcCo08, EcCo10 and EcRe11
criteria. The latter three are respectively related to benefits and local and regional
employment. In the postulate matrix, the ScPa03 and ScVa06 criteria receive a score of 4 to
indicate a maximum benefit, as in the previous postulate. The ScPa05 criterion also receives
the maximum grade since the future of the landscape is taken into account in the best possible
way in every scenario. Compared with the NOri matrix, in the NMax-NPro matrix, the grades
of the benefits and employment criteria are given a proportional increase to take into account
the results of potential negotiations between the municipality and its private partner.
Although other percentage increases could have been considered, we arbitrarily chose 10% to
illustrate the case.
98
Table 4.6 Combined evaluation matrix for the grading of the postulate Type of
ownership of wind farm – Public – private partnership
CRITERIA
ScSS01 ScSS02 ScPa03 ScPa04 ScPa05 ScVa06 ScNA07 EcCo08 EcCo09 EcCo10 EcRe11 EnPh12 EnBi13
Scenario 1 158 150 4 2 4 4 159 153,45 329 176,0 108,24 80 460
Scenario 2 0 78 4 3 4 4 84 95,70 175 85,8 56,32 0 220
Scenario 3 0 78 4 3 4 4 108 158,40 175 116,6 112,64 0 220
Scenario 4 0 0 4 4 4 4 0 0 0 0 0 0 0
4.4 Discussion
4.4.1 Criteria
Several assessments of this matrix were made by changing the grades of criteria
ScPa03 and ScVa06, by assuming that the grades of scenarios 2 and 3 (participatory wind
farm) were always more advantageous (grade equal to 4) than that of scenario 1 (promoter
wind farm, grade equal to 1), by assigning to scenario 4 (no wind farm construction, grade
99
equal to 0) the worst grade, that of not using any communication. During these evaluations,
in all cases, although the net flux values varied, the ranking results were the same: scenario 3
(build the wind farm, participatory wind farm) represented the best option. It can be noted
that the use of communication tools contributes to the implementation of wind farm.
Situation in which the criteria receive the same grade, NMax matrix
All the criteria that are likely to be changed (ScPa03 and ScVa06) receive the same
maximum grade, which is defined in a previously constructed scale (see Vazquez et al.
2013b). Here, this grade shows that significant community spaces, as well as the need to
communicate and be heard, are taken into consideration, without differences in each of the
four scenarios. We can see that the ranking of the scenarios shows that the improved grades
of the social criteria favor the implementation of a wind farm, i.e. the construction of a
participatory wind farm rather than no construction.
Moreover, if we modify this matrix by excluding the criteria that are likely to change,
the new ranking obtained will be the same as the one obtained when these criteria are
included: therefore, these criteria do not change the result, because they are all equally
effective. This result is due to the fact that the concerns expressed by the stakeholders are
taken into account. These criteria that do not have an impact on the scenario rankings
obtained with the PROMETHEE methodology are therefore considered to be non-
discriminating. Thus, in our example, we note that the maximum grade makes it possible to
evaluate what happens if the constraints related to the Significant places and Taking into
account the population’s needs and values criteria are removed:
When selecting the site where the wind turbines are to be installed, it is no longer necessary to
consider this postulate because this issue has already been resolved. As a result, the original definition
problem has changed and it is not longer necessary to answer the question of the use of communication
tools to analyze the criteria that are likely to be changed.
The highest ranked scenario is the participatory construction of a wind farm.
100
The ranking of the scenarios remains the same: the new matrix does not provide any
new information.
Here again, the ranking obtained shows that the improved grades of the social criteria
support the implementation of the wind farm: scenario 3 ( participatory construction wind
farm) represents the best option, while in the original matrix, the construction and the non
construction of the wind farm received the same grade. The final grade obtained (net flow) is
a combination of maximum grades (value equal to 4) for the Usual function and of
proportional grades, through an increase of 10% for the V-Shape functions.
4.4.2.1 Net flow between the rankings of the NOri and NMax matrices
If we eliminate from the evaluation matrices the criteria that have received the
maximum grade, the ranking of two matrices will be equal (flow: -0.28, 0.02, 0.18 and 0.08),
and this is the case for the Communication tools postulate and its Negotiation table sub-
postulate as well as for the Current situation postulate, since both matrices will be the same.
In addition, in the evaluation of the NOri matrix of the Current situation postulate, if
we remove the non significant criteria identified in NMax matrix, the original ranking (flow:
-0.40, -0.02, 0.21 and 0.21) changes and the grades of one of the wind farm construction
options (scenario 3, participatory wind farm, flow = 0.18) and of the no construction option
(scenario 4, flow = 0.08) are no longer equal. Indeed, a construction option, the one of the
participatory scenario, is clearly preferred and becomes, according to the calculations, the
best option. We note that:
101
Based on the flow results, to address strategic issues upstream of the decision-making
process in a wind farm promotes its execution: flow equal to 0.18 for the construction
(scenario 3) compared with a flow of 0.08 with no construction (scenario 4). The Significant
places and Population’s needs and values criteria analyzed here are related to access to land
rights that people had before the existence of the wind farm. For example, interim control
regulations could have resolved these problems. If these issues are not addressed early on the
decision process, they reappear in the construction stage of a given project (Côté 2011).
So, no answer is given to these questions, we are faced with a negotiation: in the
NOri evaluation matrix the resulting flows are equal to 0.21 for both the construction
(scenario 3) and the no construction (scenario 4) options. This will require the identification
of criteria to be negotiated and, in our case, these would be non discriminatory criteria
identified for the Communication tools postulate.
If there is no answer to the strategic questions upstream of the decision-making
process and if we remove from the NOri matrix (Current situation postulate) the criteria
ScPa03 (Significant places) and ScVa06 (taking into account the population’s values and
needs) the ranking changes and the construction of the wind farm is inaccurately identified as
the best option (flow = 0.18). However, this will not identify the improvements that address
the concerns of the population, due to the loss of one of the advantages of the MCDA, that of
the negotiation process.
This shows that if the strategic question of the communications tools comes back to the
negotiation table, the use of such tools would further encourage the implementation of the
wind farm.
4.4.2.2 Net flow between the NOri and NMax-NPro matrix rankings
As noted earlier, the proportional changes made to the criteria belonging to Usual
function do not change the result, unlike those made on the V-Shape functions in which a
10% increase changes the proportion of the performance difference between the scenarios
(actions). This change in the differences affects the linear V-Shape function, and produces
different net flows in the NOri (-0.40, -0.02, 0.21 and 0.21) and NMax – NPro (-0.24, 0.01,
0.18 and 0.05) matrices.
102
4.4.2.3 Net flow between the NDif and NMax matrix rankings (Communication tools
postulate)
First, it is necessary to consider that scenarios 2 and 3 already had the best grade (4) in
both matrices: since the beginning, these scenarios involved a communication process. As for
scenarios 1 and 4 of the NDif matrix, we consider that this communication process did not
exist or was incomplete. This is why we need to understand the new grade (4) of scenario 1
of the NMax matrix as the result of a promotion of discussions on the proposed initial wind
farm proposal, while the new grade (4) of scenario 4 of the same matrix represents the
increased discussion on the issues related to the non construction of the wind farm.
We then target the discussions on the conflicting elements, i.e. the best and worst
options and on whether or not the construction wind farm is carried out. In both these
matrices, scenario 3 is the best option while scenario 1 is the worst.
Scenario 3 represents the possibility of going forward with the construction of the wind
farm (participatory wind farm) and scenario 4, no construction of the wind farm. The NDif
matrix rankings (see Table 4.4) indicate a greater difference between the best and the worst
options (interval difference between -0.32 and 0.26 = |58|), compared with that of the NMax
matrix (see Table 4.5) (interval difference between -0.28 and 0.18 = |46|). In addition, there is
a greater difference between the construction (scenario 3) and the non construction of the
wind farm (scenario 4) of the NDif matrix (interval difference between 0.26 and -0.04 = |30|)
compared to the NMax matrix (interval difference between 0.18 and 0.08 = |10|). These two
differences are therefore greatest in the “unimproved” NDif matrix.
On the one hand, the reduction in the difference between the best and the worst options
of the NMax matrix (|46|) is explained by the fact that the grades of the two likely criteria
(ScPa03 and ScVa06) of scenario 1 (worst option) were improved, although in scenario 3
(best option) the grades of these same criteria did not change. On the other hand, the
reduction of the difference (|10| for NMax) between the construction (scenario 3) and the no
construction (scenario 4) options of the wind farm can be explained by fact that the grades of
those likely criteria improved for scenario 4 in the NMax matrix, while the grade for scenario
3 remained the same. These grade changes for some of the criteria of some scenarios cause a
103
net flow redistribution: the scenarios that improved their grades earn points (+0.04 and +0.12
for scenarios 1 and 4, respectively) while those who do not undergo any change lose points (-
0.08 and -0.08 for scenarios 2 and 3, respectively). These «won» or «lost» points (i.e. +0.16
[+0.04 and +0.12] and -0.16 [-0.08 and -0.08]) have no effect on the net flow of NMax
matrix.
Despite losing points in the NMax matrix, scenario 3 continues to be ranked as the best
option in both matrices. Therefore, in the NMax matrix, the worst option (scenario 1) and the
no construction option (scenario 4) improved their net flows, causing a decrease in their
differences between the best and the worst option and between the construction or no
construction scenarios (|58| vs. |46| and |30| vs. |10|). In this way, communication, which is at
the base of the postulate analyzed above, was used to improve the wind farm (scenarios),
while maintaining scenario 3 as the best option.
4.4.2.4 Net flow between the NMax and NMax-NPro matrix rankings.
Once again, we focus the discussion on the conflicting elements (best and worst option
/ construction of the wind farm or no construction), this time by modifying social and
economic criteria since this is a requirement of the Type of ownership of wind farm postulate.
In this case, the interval difference between the best and worst options (interval difference -
0.28 and 0.18 = |46|) of the NMax matrix (see Tab. 8.5) is equal to that of the NMax-NPro
matrix (see Tab. 8.6) (interval difference -0.24 and 0.18 = |42|). The best option continues to
be scenario 3 and the worst, scenario 1. The difference between building the wind farm
(scenario 3, participatory wind farm) and not building it (scenario 4) in NMax matrix
(interval difference 0.18 and 0.08 = |10|) is lower than that of the NMax-NPro matrix
(interval difference 0.18 and 0.05 = |13|). We note here, for the NMax-NPro matrix, that there
is a decrease in the difference of deviation between the best and the worst option (|46| and
|42|), and at the same time there is an increase in the difference of deviation between building
and not building the wind farm (|10| and |13|).
In the redistribution of the total flows of this example, scenario 3 is unchanged and
retains a flow equal to 0.18. The three remaining scenarios undergo changes with either an
increased or decreased flow. This explains the increase or the reduction of the difference
104
shown above. In the NMax-NPro matrix, the scenarios whose grades for the likely criteria
have improved the most get points (+0.04 for scenario 1) while the scenarios that keeps the
same grades as those of the NMax matrix loses points (-0.01 for scenario 2 and -0.03 for
scenario 4).
Just like in the previous case, the points that are “won” or “lost” (i.e. +0.04, -0.01 and -
0.03) have no effect on the net flow of the NMax-NPro matrix. Scenario 3 continues to be
ranked as the best option. Especially in the NMax-NPro matrix, the improvement of the net
flow of the worst option (scenario 1) (+0.04) is greater than that of the best option (0). This
explains the decrease in the difference between these two options (|46| and |42|) of the two
matrices analyzed here. The no construction scenario has a reduced net flow because it does
not contribute anything to the job creation and local financial benefits criteria, while the
construction of the wind farm does improve the results of these criteria. The grades of the
other three scenarios are increased by 10%. This is why the difference between the scenarios
“construction and no construction of the wind farm is greater in the NMax-Npro matrix (|13|)
than in the NMax matrix (|10|), a result that supports even more the possibility of building the
park.
No robustness analysis of the Taxation and royalties postulate was done because it has
a reduced number of likely criteria. This is why this postulate was chosen as an example for
discussion on the redefinition of the problem. This postulate has been exclusively designed to
highlight the current benefit distribution landscape of wind farms. It is not an analysis of the
effects of increased taxation or royalties of wind farms here in Quebec or elsewhere (i.e. en
France). Neither is it intended as a proposal. It is simply designed to logically structure
concerns expressed by stakeholders in order to respond to the strategic question regarding the
redistribution and sharing formulas of revenues generated by wind farms.
If we consider that these formulas are based on a regulatory framework, setting a tax
could be considered as an option. If such a tax existed in Quebec, and depending on the
concerns expressed by the stakeholders, the sharing of benefits could be done according to
one of the following five formulas:
105
Although these sharing formulas structure and represent the stakeholders’ concerns,
they are not the only possible formulas. To identify them, new analyses would be required in
which other factors such as the maximum amount of community benefits would be defined
by the wind farm promoter. Such an amount could be based on a cost-benefit study. Other
considerations could include a special tax regime for wind farm developers. Such a tax
regime would need to take into account its impacts on the local benefits and the actual
conditions required for it to really produce benefits for the community.
4.6 Conclusion
Firstly, the comparison between the Current situation with Communication tools or
Type of ownership of the wind farm, made it possible to identify objective rules based on
changes in the ranking of scenarios. This change was verified through a robustness analysis
using evaluation matrices containing different, maximum and proportional values with
respect to the values of the original matrix. Nevertheless, this analysis always involves a
subjective aspect. This subjectivity comes from the values of the qualitative scale of some of
the criteria indicators that were used to build the new evaluation matrices.
The observed relationships between the results of the different postulate matrices lead
to the following rules (also represented in the diagram below, see Fig. 4.6):
1) Same criteria and same evaluation matrix: nothing changes (it is not
necessary to redefine the problem) (NOri Matrix).
2) Same criteria and a matrix with proportional improvements (constant
change): the ranking will be the same (a proportional improvement of the criteria grades does
not affect the ranking) (NPro matrix).
106
Secondly, to achieve this robustness analysis, three evaluation matrices were built.
However, several others could have been designed as, for example, the 24 listed in Tab. 4.1.
The number of matrices could have increased because it would have been necessary to add
matrices with different, maximum and proportional grades, which would have multiplied the
total number of matrices. This is why, to quickly carry out a robustness analysis on a large
number of matrices reflecting several strategic postulates, it is necessary to use automation in
the corresponding module of the software.
Finally, as noted earlier, the strategic postulates were developed in 2011, by structuring
the concerns expressed in public documents submitted to the BAPE in 2005. This was done
on the basis that there are concerns not only about where the turbines will be located but also
about the strategic planning of wind farm. The design of such postulates must adequately
reflect the issue because it can bring to light several possibilities that could warrant an
analysis. The design of such postulates is still relevant today because, for some stakeholders,
strategic concerns continue to be at the center of public debate, as shown in the last public
report of the BAPE16 related to a wind farm.
Figure 4.6 Rules followed for the changes in the scenario rankings
16
Public Hearing Report 288, July 2012 [online]
Available at http://www.bape.gouv.qc.ca/sections/rapports/publications/bape288.pdf
(Accessed October 2nd, 2012)
108
4.7 References
Buuren J.C.L. van and Hendriksen A (2010) ‘A Learning and Decision Methodology for
Drainage and Sanitation Improvement in Developing Cities’. Dans Social Perspectives on the
Sanitation Challenge, sous la direction de Vliet, B., Spaargaren, G. et Oosterveer, P.
Dordrecht, Springer Verlag, 2010. New York
Corsair HJ, Bassman J et al. (2009) Multicriteria Decision Analysis of Stream Restoration:
Potential and Examples. Group Decision and Negotiation 18(4). Springer Science Businnes
Media
Kain JH, Söderberg H (2008) Management of complex knowledge in planning for sustainable
development: The use of multi-criteria decision aids. Environmental Impact Assessment
Review 28(1): 7-21
Laaribi A, Chevallier JJ, Martel JM (1996) A spatial decision aid: a multicriterion evaluation
approach. Comput., Environ. And Urban Systems, 20(6):351-366
Longley PA, Goodchild M, Maguire DJ, Rhind D (2011) Geographic Information Systems
and Science. John Wiley&Sons, Inc. New York
Lovett A, Appleton K (2008) GIS for Environmental Decision-Making. CRC Press. Taylor &
Francis Group
Malczewski J (1999) GIS and multicriteria decision analysis. John Wiley & Son, Inc. New
York.
110
Nobre A. et al (2009) Geo-spatial multi-criteria analysis for wave energy conversion system
deployment. Renewable Energy 34: 97-111
Prades JS, Loulou R, Waaub JP (1998) Stratégies de gestion des gaz à effet de serre. Le cas
des transports urbains. Presses de l’Université du Québec
Roy B (2002) Robustesse de quoi et vis-à`vis de quoi mais aussi robustesse pourquoi en aide
à la décision? Université Paris-Dauphine. EWG-MCDA Newsletter, Autumn
Vazquez M de L, Waaub JP and Ilinca A (2013a) « Coupling MCDA and GIS in a Decision
Making Process for Wind Farm Projects Analysis –TIMED Approach. » International
Journal of Geomatics and Spatial Analysis, VOL 23/1 - 2013 - pp.95-123 -24.
doi:10.3166/rig.23.95-123. Lavoisier.
Vincke P (1999) Robust Solutions and Methods in Decision-Aid. J. Multi-Crit. Decis. Anal.
8: 181–187
CHAPITRE V
Après le développement du cadre théorique de DEMIT, nous avons testé l’approche à l’aide d’une étude
de cas. La réussite de la prise de décision participative est conditionnelle à la construction de quatre espaces :
physique, intellectuel, émotionnel et procédural. Nous démontrons dans cet article par une analyse FFOC (forces,
faiblesses, opportunités, et contraintes) (en anglais l’acronyme SWOT est utilisé: Strengths, Weaknesses,
Opportunities, and Threats) que le bilan forces-faiblesses et opportunités-contraintes de DEMIT est positif et que
ces forces deviennent les vecteurs d’une décision participative. Ceci est réalisé par l’interaction accrue entre les
connaissances scientifiques et locales, et par la reconnaissance de l’importance des systèmes de valeurs dans
l’analyse socioéconomique, culturelle, et environnementale des enjeux d’un projet.
17
M. de L. Vazquez ()
Doctorante en sciences de l’environnement, Laboratoire de recherche en énergie éolienne, Université du Québec à
Rimouski (UQAR). 300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
courriel: vazquez_rascon.maria_de_lourdes@courrier.uqam.ca
18
Ph.D., Professeur, département de géographie, Groupe d’études interdisciplinaires en géographie et environnement
régional, Groupe d’études et de recherche en analyse des décisions. Université du Québec à Montréal (UQAM), Case
postale 8888, succursale Centre-ville, Montréal, H3C 3P8, Qc, Canada
courriel: waaub.jean-philippe@uqam.ca
19
Ph.D. Professeur, département de mathématiques, informatique et génie, Laboratoire de recherche en énergie
éolienne, Université du Québec à Rimouski (UQAR) 300, allée des Ursulines, Rimouski, G5L 3A1, Qc, Canada
courriel: adrian_ilinca@UQAR.QC.CA
112
5.1 Introduction
Le but de cet article est de vérifier dans quelle mesure l’utilisation de l’approche
DEMIT réunit les conditions physiques, intellectuelles, émotionnelles et procédurales
nécessaires à une prise de décision participative et transparente.
alternatives possibles, information reliée à des faits concrets obéissant à une réalité concrète.
De ce fait, l’espace intellectuel a un caractère objectif (Stanfield, 2000).
Par ailleurs, dans le débat, l’échange d’informations portant sur une réalité extérieure
et objective interpelle également le système de valeurs des acteurs en présence, suscitant de
leur part une réponse interne (c.-à-d. association d’images et/ou d’émotions) (Stanfield, 2000)
qui agit sur l’espace émotionnel. Cet espace émotionnel est aussi lié au contexte
d’interprétation des interrelations psychosociales entre les acteurs participants au processus.
L’établissement d’une procédure est indispensable pour arriver à une prise de décision
(Sintomer et al., 2012). Cette procédure est définie par des étapes à suivre pour une prise de
décision participative. En ce qui concerne DEMIT, il s’agit par exemple de l’identification du
problème, la construction de critères et indicateurs, la construction des alternatives, etc.. De
plus, dans cette procédure, il faut considérer l’existence d’un certain cadre réglementaire qui
pourrait avoir une incidence sur elle (exemple relatif à DEMIT: règlement de contrôle
intérimaire) et tenir compte des conditions particulières auxquelles une étape pourrait être
assujettie (exemple relatif à DEMIT: manque de participation des acteurs, information
confidentielle, etc.).
115
5.2 Méthodologie
Les grandes étapes de l’analyse FFOC sont : la détermination des conditions générales
de l’analyse, la préparation des réunions, l’analyse des quatre facteurs, la mise en relation des
facteurs, la synthèse de l’analyse et la production de graphiques pour mettre en évidence les
principaux résultats de façon logique. Ces étapes seront suivies lors de l’analyse FFOC de
l’approche DEMIT pour déterminer, de façon logique et rapide, dans quelle mesure DEMIT
aboutit à son objectif de favoriser une prise de décision transparente et participative. Du fait
qu’il ne s’agira pas d’une analyse complexe, elle comprend seulement quatre espaces et deux
modules, cette analyse FFOC se fait sans la coupler à une autre méthode.
avec les outils multicritères AMCD et SIG (voir Vazquez, Waab et Ilinca, 2013b, section 2.3
et section 2.4).
Ainsi, bien que l’analyse FFOC soit réalisée exclusivement sur les deux modules de
décision multicritère (AMCD et SIG), les quatre modules proposés par DEMIT (AMCD,
SIG, CS-CL et ICA) y sont inclus. L’analyse FFOC a donc deux unités d’analyse tout en
abordant les quatre modules.
La conception de la grille d’analyse FFOC est basée sur l’étude de cas réalisée pour
tester l’approche DEMIT (voir Vazquez, Waaub et Ilinca, 2013b). Cette grille répond aux
questions spécifiques posées sur l’efficacité de cette approche pour une prise de décision
participative et transparente.
La grille développée a une forme matricielle (Tableau 5.1). Elle comprend deux
sections : une horizontale comportant les quatre espaces favorisant la prise de décision
participative et transparente, l’autre verticale incluant les quatre facteurs de l’analyse FFOC :
forces-faiblesses pour l’environnement interne, opportunités-contraintes pour
l’environnement externe. Le module assujetti à l’analyse (soit SIG ou AMCD) ainsi que
l’étape du processus sont indiqués, tels que décrits dans la section suivante. Les totaux des
forces, faiblesses, opportunités et contraintes identifiées, sont calculés et servent à construire
un tableau synthèse. Finalement, au bas de la grille, les besoins relatifs à la valorisation des
forces et des opportunités et les défis relatifs aux faiblesses et aux contraintes qu’il faut
surmonter, pour améliorer la construction des espaces garantissant une décision participative
et transparente, sont indiqués.
Sur la section horizontale, constituée des axes, on trouve quatre espaces : les trois
premiers correspondent aux aspects physique, intellectuel et émotionnel et appartiennent à
une des approches éprouvées pour réussir à construire des points communs (Delaney, 2010);
le quatrième et dernier espace fait quant à lui référence à la procédure nécessaire à
l’établissement d’une approche participative (Sintomer et al., 2012).
118
Sur la section verticale, on trouve les forces, les faiblesses, les opportunités et les
contraintes, que nous appelons facteurs. Ceux-ci sont divisés en facteurs internes et externes,
et serviront à évaluer la réussite (ou non) de la construction des axes (espaces). Les facteurs
internes correspondent aux forces et faiblesses rencontrées à l’intérieur de l’application de
l’approche DEMIT pour construire lesdits espaces (physique, intellectuel, émotionnel et
procédural). Ces forces et faiblesses proviennent des conditions spécifiques auxquelles sont
assujetties les acteurs participants (ex. : disponibilité, représentativité, etc.) et l’équipe de
soutien spécialisée en aide multicritère à la décision (ex. : multidisciplinaire ou non,
ressources disponibles, etc.). Les facteurs externes font référence aux conditions,
opportunités et contraintes, que ni l’équipe de soutien spécialisée en AMCD, ni les acteurs
participants ne peuvent contrôler (ex. : l’application d’une loi, l’information disponible,
l’avancement des connaissances, etc.).
Forces
Faiblesses
EXTERNE
Opportunités
Contraintes
TOTAL
BESOINS/DÉFIS
119
Une grille d’analyse a été conçue pour chacune des deux méthodes de décision
multicritère : SIG et AMCD. La partie relative aux SIG comporte cinq étapes, tandis que
celle concernant l’AMCD en compte 11. Les facteurs (forces-faiblesses, opportunités-
contraintes) de chaque étape sont notés sur une feuille de travail spécifique à l’étape en
question. Il y a donc un total de 16 feuilles de travail (voir annexe F). De plus, l’annexe G
présente la liste détaillée des facteurs décelés contenus dans ces 16 feuilles de travail.
Le logiciel supportant ce module a été ArcGIS, mais nous avons fait des remarques au
sujet de gratuiciels de SIG (notamment SavGis).
Le logiciel supportant ce module a été D-Sight, mais il existe des gratuiciels d’AMCD
(notamment Visual Prométhée).
Chacune des étapes a été codifiée tel qu’indiqué au Tableau 5.2. Le code a été conçu à
partir des trois premières lettres significatives de l’étape du module auquel elle appartient :
SIG pour les systèmes d’information géographique, MC pour l’aide multicritère à la décision.
Tableau 5.2 Assignation des codes servant à la description de chacune des étapes
Suite à la conception de la grille pour les deux modules SIG et AMCD, l’équipe de
soutien spécialisée en AMCD a réalisé une première analyse FFOC. Celle-ci a été soumise à
un groupe d’experts pour démarrer la discussion. Originalement, 130 facteurs ont été
présentés au groupe d’experts. Via l’implémentation d’un groupe de discussion, les experts et
l’équipe d’AMCD ont enrichi l’analyse préliminaire pour procéder à sa validation finale.
121
Nous avons mis en place un groupe de discussion pour valider l’analyse FFOC de
chaque module multicritère, SIG et AMCD. Huit experts, divisés en deux groupes (4 pour le
SIG et 4 pour l’AMCD) ont interagi pour répondre à la question principale sur la pertinence
et l’enrichissement des forces-faiblesses-opportunités-contraintes décrites sur les 16 feuilles
de travail de la grille d’analyse.
Dans un premier temps, nous avons identifié les éventuels experts participants. Les
critères de sélection ont été : emplacement géographique (centre-ville), nature de l’expertise
et interdisciplinarité. Ensuite, chaque expert identifié a reçu une invitation à participer dans
laquelle l’objectif de l’exercice était expliqué : la validation d’une analyse FFOC d’une étude
de cas réalisée pour tester l’approche DEMIT. Suite à l’acceptation de l’invitation et au
consentement à participer tout en ayant la garantie de l’anonymat des résultats (protocole
éthique), les équipes de discussion ont été constituées pour chaque module.
Une fois finalisée la constitution des équipes d’experts, nous avons planifié des
réunions de discussion, chacune d’une durée de deux heures. L’équipe qui validait le module
SIG s’est réunie une seule fois pour compléter cinq feuilles de travail. En raison de la
nécessité manifeste de disposer de plus de temps pour la réflexion, une fois la réunion
réalisée, le travail a été poursuivi par messagerie électronique. L’équipe qui validait le
module AMCD s’est réunie deux fois pour compléter 11 feuilles de travail. Finalement, tous
les commentaires reçus, autant lors des réunions de discussion que par messagerie
électronique, ont été intégrés pour en obtenir la validation finale des 16 grilles d’analyse
FFOC des modules SIG et AMCD (voir annexe F).
Le travail réalisé par l’équipe d’AMCD et par le groupe d’experts a abouti au repérage
de 230 facteurs associés aux étapes de l’approche DEMIT, dont 144 appartenant à son
environnement interne et 86 à son environnement externe (Tableau 5.3). Grâce au groupe de
discussion, 100 facteurs se sont rajoutés aux 130 initialement proposés, montrant
l’importance de compter sur une équipe multidisciplinaire et d’experts chevronnés pour
valider l’analyse.
Ces 230 facteurs ont été identifiés grâce aux compétences des participants, ce qui est
une caractéristique de l’analyse FFOC (Kurttila et al., 2000). L’importance donnée à ces
facteurs provient d’une analyse qualitative. Toutefois, ces facteurs sont considérés comme
stratégiques pour l’avenir de la construction des espaces inclus dans l’approche DEMIT. Ces
espaces faciliteront l’atteinte des objectifs en s’appuyant sur les points forts identifiés à
123
l’interne et sur les opportunités repérées à l’externe. Les faiblesses internes et les contraintes
externes affectant les méthodes multicritères doivent être connues pour minimiser leur
influence.
Le tableau synthèse 5.3 a été constitué pour décrire les résultats de la validation de
l’analyse FFOC. Il contient tous les éléments de la prise de décision participative et
transparente, la division des environnements interne et externe, les forces-faiblesses et les
opportunités-contraintes pour chacune des étapes selon les quatre espaces à construire. Pour
établir un lien visible entre les facteurs et indiquer les différences, un code couleur a été
établi : vert pour les forces, rouge pour les faiblesses, bleu pour les opportunités et orange
pour les contraintes. Pour faciliter l’analyse des résultats, le sous-total par catégorie est
indiqué.
PHYSIQUE 1 3 2 2 2 0 0 1 0 0 0 0 0 0 0 0 11
INTELLECTUEL 4 5 2 1 1 8 2 5 3 3 3 1 4 3 7 4 57
FACTEURS INTERNES
FORCES
ÉMOTIONNEL 2 1 2 1 3 2 2 0 1 2 1 1 3 2 3 2 28
PROCÉDURAL 2 1 2 1 1 1 1 1 1 1 3 1 1 1 2 1 21
SOUS-TOTAL 9 10 8 5 7 11 5 7 5 6 7 3 8 6 12 7 116
PHYSIQUE 1 2 2 3 1 0 1 0 0 0 0 0 0 0 0 0 11
INTELLECTUEL 2 1 3 1 1 1 1 0 0 0 0 0 0 0 0 0 11
FAIBLESSES
ÉMOTIONNEL 2 1 0 1 1 0 0 1 1 1 0 0 0 0 0 0 9
PROCÉDURAL 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
SOUS-TOTAL 5 4 5 5 3 1 2 1 1 1 0 0 0 0 0 0 28 144
PHYSIQUE 1 3 4 1 4 1 0 0 1 1 0 0 0 1 1 1 19
INTELLECTUEL 1 1 2 1 0 3 2 1 1 1 1 3 0 1 1 1 20
FACTEURS EXTERNES
OPPORTUNITÉS
ÉMOTIONNEL 1 1 0 1 0 0 0 0 0 0 0 0 0 1 0 1 5
PROCÉDURAL 1 0 0 1 1 2 0 1 0 1 0 0 0 1 1 1 10
SOUS-TOTAL 4 5 6 4 5 6 2 2 2 3 1 3 0 4 3 4 54
PHYSIQUE 1 4 4 0 2 0 0 0 1 0 0 0 0 0 0 0 12
INTELLECTUEL 1 1 1 0 1 6 2 0 0 0 0 0 0 0 0 0 12
CONTRAINTES
ÉMOTIONNEL 1 1 0 0 0 2 0 0 0 0 0 0 0 0 0 0 4
PROCÉDURAL 1 0 0 1 0 1 0 0 1 0 0 0 0 0 0 0 4
SOUS-TOTAL 4 6 5 1 3 9 2 0 2 0 0 0 0 0 0 0 32 86
SOUS-TOTAL par feuille 22 25 24 15 18 27 11 10 10 10 8 6 8 10 15 11
SOUS-TOTAL par module 104 126
TOTAL 230
Tableau 5.3 Analyse FFOC de la construction des espaces nécessaires à une décision participative et transparente
lors de l’application de l’approche DEMIT.
125
140
130
120
116 Vue d'ensemble
110
100
90
80
70
60 54
50
40 28 32
30
20
10
0
Forces Faiblesses Opportunités Contraintes
ENVIRONNEMENT INTERNE ENVIRONNEMENT EXTERNE
Cela dit, l’histogramme montre, pour l’environnement interne, que le module AMCD
compte 38 forces de plus que le module SIG. Par contre, les faiblesses du module AMCD (6)
sont moindres que celles du module SIG (22). Donc, le module AMCD a plus de forces et
moins de faiblesses que le module SIG.
contraintes que le module AMCD lors de la construction des espaces nécessaires à la décision
participative et transparente.
90
80 77 Vue d'ensemble
70
60
Faiblesses
50 Opportunités
39
40 Contraintes
30 Forces
30 24
22
19
20
13
6
10
0
MODULE SIG MODULE AMCD
reliées au contexte externe (9). C’est pourquoi, il faut miser sur les opportunités (6). Le fait
qu’autant de facteurs soient en compétition dans cette étape indique la nature essentielle de la
définition du problème dans la recherche d’une décision participative. L’étape EchMC12
(Choix des échelles d'évaluation, module AMCD) comprend seulement trois forces, mais elle
compte aussi un nombre équivalant d’opportunités (3).
InfSIG1
DecMC16 12 SupSIG2
10
JugMC15 DonSIG3
8
6
PreMC14 SceSIG4
4
2
MatMC13 0 CarSIG5
EchMC12 ProMC6
FORCES
IndMC11 EnjMC7
FAIBLESSES
PonMC10 MetMC8 OPPORTUNITÉS
CriMC9 CONTRAINTES
Nous avons représenté séparément les faiblesses internes et les contraintes externes
pour raffiner l’analyse. Le graphique radar des faiblesses est présenté à la figure 5.4. Le
graphique montre que la plupart des faiblesses de DEMIT se concentrent dans le module SIG,
et sont proportionnellement réparties dans ses cinq étapes, pour un total de 22 faiblesses.
Quant au module AMCD, il compte peu de faiblesses, lesquelles sont concentrées dans les
cinq premières étapes. C’est l’étape EnjMC7 (Structuration des enjeux, module AMCD) qui a
le plus grand nombre de faiblesses dans le module AMCD, ce qui indique le besoin de
trouver des mécanismes pour surmonter les limitations de cette étape. Par ailleurs, les
dernières étapes du module AMCD n’ont aucune faiblesse. Ce module a un total de 10
faiblesses, réparties sur six de ses 11 étapes, soit sur 54,5 % du total.
FAIBLESSES
InfSIG1
DecMC16 10 SupSIG2
JugMC15 8 DonSIG3
6
PreMC14 4 SceSIG4
2
MatMC13 0 CarSIG5
EchMC12 ProMC6
IndMC11 EnjMC7
FAIBLESSES
PonMC10 MetMC8
CriMC9
Figure 5.4 Graphique radar du nombre de faiblesses par étape de l’approche DEMIT
Le graphique radar des contraintes est présenté à la figure 5.5. Dans le module SIG,
l’étape SupSIG2 (Support informatique - ordinateur et logiciel) présente le plus grand
nombre de contraintes, et dans le module AMCD, c’est l’étape ProMC6 (Définition du
problème et identification des enjeux). Le module SIG présente des contraintes à chacune de
ses étapes, soit dans 100 % du module, pour un total de 19. Le module AMCD présente des
contraintes réparties dans trois de ses 11 étapes, soit 27,3 %, pour un total de 13. Le module
129
CONTRAINTES
InfSIG1
DecMC16 10 SupSIG2
JugMC15 8 DonSIG3
6
PreMC14 4 SceSIG4
2
MatMC13 0 CarSIG5
EchMC12 ProMC6
IndMC11 EnjMC7
CONTRAINTES
PonMC10 MetMC8
CriMC9
Les graphiques précédents ont permis d’analyser la distribution du nombre des forces-
faiblesses et des opportunités-contraintes dans les deux modules de DEMIT. La description
détaillée des modules ayant les facteurs les plus nombreux, tel qu’identifié plus bas, permet
de tirer des conclusions supplémentaires (pour la liste complète des facteurs répertoriés, voir
l’annexe G).
Dans l’environnement interne de DEMIT (voir tableau 5.4), le plus grand nombre de
forces du module SIG se retrouve dans l’étape SupSIG2, tandis que, pour le module AMCD,
ce sont les étapes ProMC6 et JugMC15 qui en contiennent le plus. Les trois étapes InfSIG1,
DonSIG3 et SceSIG4 du module SIG contiennent le plus grand nombre de faiblesses. L’étape
EnjMC7 englobe le plus grand nombre de faiblesses du module AMCD.
130
DonSIG3
Coûts associés (c.-à-d. nombre de séances pour les SIG participatifs, achat des
cartes, salaires des experts, équipe, logiciel)
Difficulté d’accès aux données
131
SceSIG4
Apparence d’objectivité pouvant brouiller les messages auprès de certains acteurs et
risques de manipulations
Coûts associés (c.-à-d. nombre de séances pour les SIG participatifs, achat des
cartes, salaires des experts, équipe, logiciel)
Volume de l’information à traiter face à la fluidité du processus participatif
ProMC6
Communication, débat, clarification d’idées
Évolution de la définition du problème
Formulation du problème
Identification de possibles blocages du processus dus à des enjeux non négociables
Implication de tous les acteurs
Interrelation des connaissances scientifiques et locales
Mise en valeur des systèmes de préférences de chaque acteur
Opportunité de réfléchir sur les impacts d’un projet
Reconnaissance des dimensions environnementale, sociale et économique des
enjeux
Reconnaissance de l’importance que ces systèmes ont dans le dialogue et la
négociation
Reconnaissance objective du contexte auquel font face les acteurs
JugMC15
Aide à identifier quels critères pourront faire l’objet de la négociation
Aide à identifier si les recommandations résultant de l’analyse sont solides ou non
132
D’une part, les facteurs détaillés ci-dessus indiquent que le module SIG de DEMIT, en
plus de contribuer avec ses forces très connues (capacité de stockage, etc.), favorise la
représentation d’idées concrètes. D’autre part, ces facteurs montrent qu’allouer des budgets
pour les coûts associés à ce module (c.-à-d. experts pour traiter les données pertinentes et non
pas celles déjà existantes), ainsi qu’un code d’éthique, garant de la neutralité professionnelle,
permettront de surmonter ces faiblesses. Quant aux forces du module AMCD, en plus de
celles déjà bien connues (c.-à-d. formulation du problème), le repérage rapide des enjeux
133
Les forces de ces deux modules constituent des atouts importants pour l’atteinte de
décisions négociées ayant une forte légitimité sociale. Pour leur part, les faiblesses indiquent
des points sur lesquels l’expert qui met en pratique l’approche DEMIT, doit faire attention
pour ne pas nuire au processus de construction des quatre espaces de participation.
Coûts associés
134
Les facteurs externes détaillés ci-dessus identifient les opportunités du module SIG
participatif pour construire les quatre espaces nécessaires à la décision participative et
transparente. La recherche, déjà reconnue comme importante pour améliorer l’état des
connaissances, constitue une opportunité, ainsi que les effets d’exemple sur de nouveaux
projets d’énergies renouvelables. Pour le module SIG, les contraintes sont reliées aux coûts,
eux-mêmes considérés comme une faiblesse, ou aux budgets pour la recherche. Les forces du
module AMCD concernant les processus participatifs déjà mis en place, représentent une
opportunité autant pour la procédure existante que sur le plan de l’information générée et
disponible. Les particularités historiques d’une région ou le refus d’outils considérés trop
techniques et éloignés des acteurs sont des contraintes associées au module AMCD. Une
formation continue tout au long du processus, qui commencerait dès le début de celui-ci,
permettrait une meilleure appropriation des outils et une meilleure adhésion au processus.
Les forces et opportunités de ces deux modules favorisent la création des quatre
espaces nécessaires à la prise de décision participative et transparente. Pour leur part, les
faiblesses et les contraintes indiquent les points ralentissant une décision concertée sur la base
de l’approche DEMIT et concernant le développement et l’exploitation des énergies
renouvelables.
Les limites de DEMIT sont identifiées en partie par l’analyse FFOC. Nous
retrouvons comme limites de cette approche: le besoin d’experts en SIG participatifs et en
AMCD, le degré de neutralité et d’honnêteté de ces experts, la capacité des experts d’en tirer
des conclusions, la conception de sites Web, le contexte politique permettant de confronter
les idées, la conjoncture démocratique, l’encadrement assurant une discussion respectueuse
entre les acteurs, l’exigence du temps alloué, l’exigence de l’information requise et de celle
disponible, la difficulté d’accès à l’information, l’intégration de personnes de tous âges et de
tous contextes culturels, la perception du territoire à l’aide de méthodes informatiques, la
politisation du processus, le fait de rendre significatif le choix de l’échelle, la présentation
visuelle pas toujours assez conviviale des résultats des analyses, les coûts associés, les défis
en communication pour éviter l’effet « boîte noire », l’interprétation adéquate des systèmes
de valeurs des acteurs par les experts, la concordance entre les systèmes de valeurs des
136
acteurs et les critères proposés par les experts, le besoin de discriminer davantage à l’intérieur
de chacune des catégories des acteurs.
5.4 Conclusion
Cependant, bien que ce bilan soit positif, des facteurs stratégiques permettent de
l’améliorer: les forces augmenteront grâce à la recherche et au développement continu dans le
domaine de l’AMCD et des SIG participatifs, grâce à une participation abondante et
contributive des acteurs interpellés par un projet en particulier ainsi que grâce à l’interaction
accrue entre les connaissances scientifiques et les connaissances locales. Les faiblesses
diminueront si le processus est mené dans le respect de tous les participants, si des budgets
sont alloués pour la prise de décision participative, si les données nécessaires sont disponibles
et de qualité, si le processus est exempt de pressions pour faire penser et agir le groupe d’une
façon prédéterminée, si les experts arrivent à présenter les informations nécessaires sous une
forme simple favorisant la fluidité, et s’il existe une reconnaissance de l’importance des
systèmes de valeurs dans l’analyse des enjeux sociaux, économiques, culturels et
environnementaux d’un projet en particulier. Les faiblesses à l’intérieur du module SIG
137
doivent être surmontées en premier puisqu’elles sont réparties dans chaque étape du
processus.
Cette analyse FFOC a été réalisée avec une grille qui a été appliquée sur les résultats
du test d’application de l’approche DEMIT, test réalisé pour un parc éolien établi au Québec.
Ainsi, certaines forces-faiblesses ou opportunités-contraintes pourraient être différentes dans
un autre contexte géographique et culturel. Il est donc nécessaire de prendre en compte les
particularités culturelles, historiques et géographiques du milieu où l’analyse FFOC est
effectuée.
telle équipe devient important et devra être déterminé en fonction des besoins du projet à
analyser. Une équipe interdisciplinaire, composée d’un nombre adéquat d’experts, permettra
une meilleure définition de départ du problème à analyser. La suite dépendra également de la
participation des acteurs qui au fur et à mesure de nouveaux projets, seront de plus en plus
aptes à contribuer du fait d’une courbe d’apprentissage social et d’amélioration continue
(ex. : communautés de pratique, réseaux sociaux, sites Internet d’échanges d’expériences
etc.).
Pour conclure, bien que l’application de DEMIT comporte des faiblesses et des
contraintes, les forces inhérentes de ses modules AMCD et SIG participatifs font en sorte que
cette approche réussit à construire les quatre espaces nécessaires à la prise de décision
participative et transparente. Dans ce sens, la capacité de DEMIT d’utiliser des méthodes
multicritère existantes pour répondre à la demande de transparence et d’implication dans la
prise de décision a été validée par cette analyse FFOC.
5.5 Bibliographie
Abelson J., Forest P.-G., Eyles J., Smith P., Martin E. et Gauvin F.-P. 2012.
« Deliberations about Deliberation: Issues in the Design and Evaluation of Public
Consultation Processes, McMaster University. Centre for Health Economics and Policy
Analysis Research Working Paper 01-04.
Avritzer, L. 2002. Democracy and the Public Space in Latin America. Princeton
University Press. USA.
139
Gao C.-Y. and Peng D.-H., 2011. « Consolidating SWOT Analysis with
Nonhomogeneous Uncertain Preference Information. » Knowledge-Based Systems. 24, 796-
808. ELSEVIER.
Görener et al. 2012. « Application of Combined SWOT and AHP: A Case Study for a
Manufacturing Firm. » Procedia Social and Behavioral Sciences. 58, 1525-1534.
ELSEVIER.
Greenbaum, T. L. 1998. The Handbook for Focus Group Research. Sage Publications,
Inc.
Kurttila, M. et al. 2000. « Utilizing the Analytic Hierarchy Process (AHP) in SWOT
Analysis – a Hybrid Method and its Application to a Forest-Certification Case. » Forest
Policy and Economics. 1; 41-52. ELSEVIER.
Markovska et al. 2009. « SWOT Analyses of the National Energy Sector for
Sustainable Energy Development. » Energy. 34; 752-756. ELSEVIER.
Rehm, B. et al. 1992. « Water Quality in the Upper Colorado River Basin. » Dans
Discovering Common Ground, sous la direction de Weisbord, M. R. Berrett-Koehler
Publishers. San Francisco.
Vazquez, M. de L., Waaub, J.-P. et Ilinca, A., 2013a. « Coupling MCDA and GIS in a
Decision Making Process for Wind Farm Projects Analysis –TIMED Approach. »
International Journal of Geomatics and Spatial Analysis, VOL 23/1 - 2013 - pp.95-123 -24.
doi:10.3166/rig.23.95-123. Lavoisier.
Vazquez, M. de L., Waaub, J.-P. et Ilinca, A., 2013c. « MCDA: Measuring robustness
as a tool to address strategic wind farms issues. » Assessment and Simulation Tools for
Sustainable Energy System, chapter 8. Fausto Cavallaro (ed.), SPRINGER.
CONCLUSION GÉNÉRALE
6.1 Contributions
prend en compte tous les acteurs impliqués dans un projet de parc éolien;
propose un cadre structuré où des préoccupations à caractère subjectif, et des
variables qualitatives, sont analysées;
utilise deux outils modernes et en constante évolution qui tiennent compte du
contexte complexe d’un projet de parc éolien, soit l’AMCD et les SIG
participatifs;
a été validée par une étude de cas avec des acteurs réels pour analyser
différents scénarios d’un projet de parc éolien et élaborer la procédure de
rangement des scénarios;
143
Entre autres, cette recherche a inclus, par le biais des postulats, des préoccupations
stratégiques au sujet des conditions du développement de la filière éolienne au Québec,
exprimées par différents acteurs lors des séances de consultations publiques.
Pour que les usagers potentiels de DEMIT aient une information supplémentaire sur
l’outil proposé, nous fournissons une analyse, validée par des experts, des forces-faiblesses et
opportunités-contraintes de l’approche en regard de sa capacité à éclairer un processus
décisionnel participatif et transparent. Suite à l’analyse, nous remarquons que les forces sont
notamment supérieures aux faiblesses, et que les opportunités sont plus importantes que les
contraintes.
144
Les travaux réalisés durant cette recherche ont été présentés dans 15 conférences
internationales dans 6 pays. Les financements obtenus auprès du ministère de Relations
internationales du Québec (MRI) ont permis d’appliquer l’approche DEMIT, par le biais de
projets pilotes, dans d’autres contextes, au Mexique et au Brésil.
6.2 Applications
DEMIT peut être utilisée pour acquérir des connaissances sur le territoire et sur la
gestion des projets s’y déroulant. Son application volontaire peut être une opportunité pour
évaluer un projet de parc éolien en amont de sa construction. Cela permettrait l’implication
de la communauté hôte dans la recherche de solutions et faciliterait l’acceptabilité sociale du
parc éolien.
Les 13 critères proposés par DEMIT pourraient s’appliquer dans d’autres projets,
propositions stratégiques ou contextes pour faciliter le démarrage rapide de la discussion. Les
critères peuvent être adaptés à un contexte particulier. Concrètement, DEMIT peut
s’appliquer à la filière microhydraulique, pour laquelle les 13 critères seraient inchangés.
Cependant, les indicateurs reliés à ces critères devraient s’adapter pour bien évaluer les
impacts sur l’environnement, sur la flore et la faune des cours d’eau ou pour tout autre impact
social, culturel, économique et technique dû à la valorisation de cette ressource. Les quatre
catégories des acteurs intervenant dans l’application de DEMIT peuvent s’élargir pour
inclure, par exemple, un type d’intervenant relié aux droits d’utilisation des terrains. En effet,
certains pays de l’Amérique Latine ont un type de propriété communautaire réglementé par
des lois spécifiques qui indiquent le protocole pour l'usufruit de la terre, parmi lesquels les
exigences relatives à sa location.
Les modules proposés par DEMIT (AMCD, SIG participatifs, CS-CL et ICA)
pourraient être utilisés de façon indépendante, si l’usager le considère utile pour un projet
spécifique. De plus, DEMIT pourrait être intégrée à d’autres projets reliés à la recherche de
solutions durables lors de l’exploitation des ressources énergétiques. Du fait que sur un
territoire donné, un projet de gestion ne peut fonctionner que si les acteurs du territoire se
l’approprient (Sébastien, 2006), DEMIT s’avère très utile dans la planification territoriale
qui s’oriente actuellement vers une dimension collaborative et participative.
145
Pour une application réussie de DEMIT dans des contextes différents de celui où
cette approche a été développée, il nous semble important de souligner les
caractéristiques suivantes, tel que rencontrées au Québec et qui pourraient nécessiter une
adaptation dans d’autres contextes : un cadre réglementaire favorisant le développement
durable et la participation; une société habituée à participer et à exprimer ses
préoccupations; un État qui favorise l’intégrité physique des citoyens; disponibilité d’une
source de financement pour la tenue de séances publiques; et des sources d’information
crédibles, accessibles et en quantité adéquate conformément aux exigences de l’analyse à
effectuer.
Limites
Les systèmes de valeurs des acteurs, même s’ils ont une place dans la construction du
cadre argumentaire, ne sont pas en soi le sujet de la thèse. Ce cadre argumentaire inclut une
connaissance qui repose sur le savoir-faire et/ou sur des raisonnements objectifs, ne
représentant donc pas des préférences personnelles reliées à des systèmes de valeurs. Cette
recherche n’analyse pas non plus les éléments de gouvernance ou les relations entre les
acteurs. Aucune restriction au sujet du droit à la participation publique n’a été observée lors
de la réalisation du test. Aucun problème de concurrence entre experts et non-experts n’a été
observé lors de la réalisation de l’étude de cas. Aucun problème à caractère politique n’a été
mis en évidence lors de l’étude de cas. Bien que nous ayons développé des indicateurs
reflétant les enjeux à évaluer, le but de cette thèse est plus large et la conception de ces
indicateurs a été restreinte à cause des délais de temps à respecter et de l’absence d’une
équipe interdisciplinaire incluant différents champs d’expertise.
compter avec des gratuiciels fiables, allouer des budgets pour la prise de décision
participative et pour la recherche dans ce domaine.
Étant donné que tous les enjeux soulevés ne pouvaient être évalués par des critères et
indicateurs quantitatifs, certaines échelles qualitatives ont été élaborées en prenant en compte
le mieux possible les arguments disponibles pour justifier le passage d’un échelon à l’autre.
Dans ces cas, les évaluations peuvent être considérées comme davantage empruntées de
subjectivité avec les conséquences potentielles sur les résultats des simulations. Par ailleurs,
il faut aussi parfois avoir des réserves concernant l’apparente objectivité des évaluations
quantitatives car justement, l’approche par indicateur indique bien des choix pouvant influer
sur les résultats obtenus (ex. : valeurs relatives ou absolues, indicateurs ne prenant que
partiellement en compte certaines dimensions du problème, par exemple les espèces rares
pour évaluer les pertes potentielles de biodiversité). Cependant, le travail réalisé cherche à
avant tout répondre de façon cohérente et rationnelle aux préoccupations, enjeux et questions
soulevées par les acteurs impliqués dans le développement des projets de parcs éoliens.
Extensions
L’approche DEMIT a été développée pour répondre aux problèmes rencontrés lors du
développement des projets de parcs éoliens. Cependant, elle pourrait s’adapter à d’autres
sources énergétiques renouvelables, telle que la biomasse (DEMIT-biomasse) ou les mini-
centrales hydrauliques (DEMIT-microhydraulique). L’approche, sans être universelle,
pourrait être généralisée à d’autres projets à développer sur le territoire, en changeant un peu
147
Une autre piste de développement est l’intégration de cette approche sur une
plateforme informatique permettant l’insertion des couches SIG comprenant toutes les
informations géoréférencées décrivant le territoire d’étude (observatoire du territoire, système
d’information sur le territoire, etc.) telles que les potentiels éolien et solaire, l’hydrographie,
les infrastructures routières, les milieux boisés, la toponymie, la localisation de lignes de
transport d’énergie électrique, les milieux sensibles, les zones protégées, la flore et la faune
protégée ou sensible, la localisation de bâtiments, l’identification d’activités sur le territoire,
etc. Ces informations favoriseraient davantage l’adaptation des critères et des indicateurs qui
sont à la base de l’approche DEMIT. L’utilisation de ces informations géoréférencées, ainsi
que celles des critères et indicateurs développés, servirait aux usagers à explorer le territoire
assujetti à des projets de parcs éoliens, à communiquer entre eux et à concevoir des scénarios
qui serviraient à réfléchir sur les impacts, tant favorables que défavorables, des projets de
parcs éoliens pour les évaluer.
148
Même si, dans des recherches futures, il fallait tenir compte de différences
générationnelles dans l’utilisation d’Internet ou des problèmes d’accès à la technologie
informatique (logiciels, ordinateurs, Internet), le Web pourrait servir à élaborer des bases de
données participatives, sous réserve de validation par une autorité reconnue, et à favoriser la
transparence des résultats obtenus. Également, DEMIT pourrait servir à la conception
d’outils, informatiques ou non, d’aide au travail collaboratif, où les spécialistes de la
dynamique de groupe seraient appelés à participer.
Même si d’autres approches couplant l’AMCD et les SIG ont été documentées depuis
1991 ou si d’autres chercheurs se sont déjà penchés sur le thème de la participation, DEMIT
a été développée pour répondre, expressément, à cette demande apparue dans le domaine des
énergies renouvelables (l’énergie éolienne) au Québec et dont différentes sources témoignent,
telles que : Le Canada Français, section Actualité, mercredi le 27 février 2008; Le Journal de
149
Québec, lundi le 9 août 2010; les conclusions des rapports du BAPE sur l’énergie éolienne
appelant à proposer un projet socialement plus acceptable ou à privilégier un référendum,
contenues dans les rapports 233, 264 et 267.
DEMIT constitue ainsi un autre apport aux travaux existants et parmi lesquels on
trouve notamment –en ordre chronologique– les auteurs suivants qui proposent différentes
façons d’approcher le volet partipatif : Leonardo Avritzer et ses travaux sur la démocratie et
l’espace public (Avritzer, 2002); Jean-Pierre Gaudin avec son travail sur les limites d’une
démocratie participative (Gaudin, 2007); Ana Simão avec son travail sur la planification et la
participation publique (Simão, A. et al., 2008); Pascal Oberti, avec son concept émergeant
d’AMCDP (Oberti, 2008); Carlo Prévil, avec son travail sur la participation publique dans la
gouvernance du territoire (Prévil, 2009) ainsi que Bernard Roy, avec le concept de la co-
construction dans la démarche de la prise de décision (Roy, 2009).
150
Dans cette recherche, nous avons fait un effort structuré pour répondre aux
préoccupations des acteurs impliqués dans des projets de parcs éoliens, préoccupations qui,
parfois, sortent du simple cadre de la localisation des turbines éoliennes, et demandent
d’aborder des questions stratégiques relatives au développement des projets de parcs éoliens
ainsi que la prise en compte des valeurs des acteurs.
PRÉCISIONS MÉTHODOLOGIQUES
Nous détaillons ici certaines étapes des modules AMCD, SIG et CS-CL (trois des
quatre modules de l’approche DEMIT) réalisées pendant le test de l’approche (objectif 2) et
non précisées dans l’article à publier (chapitre II) en raison d’un manque d’espace.
MODULE AMCD
Des précisions sur la définition du problème, la structuration des enjeux, le choix des
critères, la pondération des critères, l’identification d’indicateurs, le choix des échelles de
mesure et la matrice d’évaluation sont présentées ici.
Définition du problème : nous avons conçu une feuille de travail pour appuyer les
participants dans leur réflexion sur les sources des problèmes de développement des projets
de parcs éoliens.
Choix du poids des acteurs : les différentes propositions sur le partage de l’influence
décisionnelle entre le secteur public, les experts, la société civile et le secteur privé ont servi
de base à la discussion des participants pour arriver, finalement, à un choix commun.
Structuration des enjeux : nous avons basé notre étude de cas, entre autres, sur des
informations de la consultation publique du projet à Baie-des-Sables. Cette information a été
obtenue sur le site Web du BAPE et inclut le rapport 21720. Ce rapport inclut aussi le projet à
L’Anse-à-Valleau. Ces deux projets sont classés dans le même dossier du fait qu’ils ont été
20
http://www.bape.gouv.qc.ca/sections/rapports/tous/index.htm
152
proposés par le même promoteur. Le site Web présente les documents tout en notant cette
situation. Ainsi, les mémoires21 de chacun des projets se trouvent dans le même dossier. C’est
pourquoi, nous avons pris en compte les mémoires des projets à Baie-des-Sables et à Anse-à-
Valleau, ce qui a enrichi les enjeux à considérer.
Après avoir décelé les préoccupations contenues dans ces mémoires, et pour favoriser
l’échange et la réflexion, une liste a été soumise aux acteurs. À la suite des discussions et des
classifications, ces enjeux ont été divisés selon leurs dimensions environnementale,
socioculturelle ou économique. Le volet technique du projet de parc éolien (distances
séparatrices, réseau électrique et autres) est représenté dans la variante du projet choisie par le
promoteur et se retrouve dans le scénario 1 à évaluer.
Choix des critères : les préoccupations décelées dans les mémoires ont servi à
déterminer les critères associés aux différents enjeux, divisés selon leurs dimensions
environnementale, socioculturelle où économique. Des vérifications ont été faites pour
déterminer si des recommandations normatives ou autres existaient déjà, telles que les
distances séparatrices indiquées dans les RCI. Une fois les critères identifiés, ils ont fait
l’objet de discussions à la table de dialogue et de négociation. Si le critère est validé, le
travail se poursuit; sinon, le processus est recommencé. La structuration se poursuit le plus
loin possible pour arriver à avoir des critères bien représentatifs des préoccupations
exprimées par les acteurs impliqués dans le projet de parc éolien.
Pondération des critères : tel qu’indiqué dans Vazquez, Waaub et Ilinca (2013b), la
pondération des critères sert à établir le système de priorités de chaque acteur correspondant à
son système de valeurs. Pour ce faire, une feuille de travail a été conçue, permettant aux
acteurs d’assigner une valeur d’importance à chaque critère (voir tableau A.1).
21
http://www.bape.gouv.qc.ca/sections/mandats/eole-valleau-sables/documents/liste_documents-DT-DQ-
DM.htm#DM
153
Pondération
Dimensions Structuration des critères Critères
(%)
Impact sur la santé et la 1. Exposition des résidences
sécurité. dans le domaine du parc,
selon les distances
séparatrices autorisées
2. Diminution potentielle de
l’espérance de vie, zone
urbaine
Sociale
tourisme
10. Emplois
Sous-total
Dégradation du milieu 12. Dégradation du milieu
Environnem
physique physique
Perte de la biodiversité 13. Perte de la biodiversité
entale
Sous-total
Total 100
154
Identification des indicateurs : dans Vazquez, Waaub et Ilinca (2013b), tous les
indicateurs ont été présentés et il a été établi que la conception de ces indicateurs a été basée
sur une revue de littérature. Les lectures faisant partie de cette revue sont présentées ici.
Les indicateurs du critère Impact sur la santé et la sécurité, et ses deux sous-critères
correspondants, sont basés sur :
Wind Turbine Sound and Health Effects. An Expert Panel Review (Colby et
al., 2009)
L’indicateur du critère Prise en compte des valeurs et des besoins des populations a
demandé une réflexion basée sur la lecture des conclusions de 12 rapports du BAPE22 pour la
période du 21 février 1997 au 14 juillet 2010 :
22
Rapports pris dans la page Web du BAPE : http://www.bape.gouv.qc.ca/.
156
Les lectures sur lesquelles nous nous sommes basées pour l’indicateur du critère
Perception de la non-acceptabilité sociale sont :
Les textes suivants ont servi à aborder les indicateurs sur les critères en lien avec les
impacts économiques dans la communauté hôte et dans la région :
23
Réalisée par l’Institut de recherche et de développement en agroenvironnement. Recueil de cartes du
Québec, 1977. http://www.irda.qc.ca/_ftbFiles/documents%20sur%20les%20sols/Carte5_categ.pdf
157
Choix des échelles de mesure : ces échelles ont été construites à partir d’une analyse
spatiale (échelle cardinale) combinée à un tableau Excel. De plus, une valeur numérique a été
attribuée pour indiquer un ordre de classement (échelle ordinale). Ces types d’échelles de
comparaison sont utilisés dans l’aide à la négociation multicritère (Maystre et Bollinger,
1999).
Pour les échelles ordinales, la revue de littérature indique les échelles suivantes :
Safe zone, low risk zone, high risk zone (Manzurul Hassan et al., 2003).
24
http://www3.mrnf.gouv.qc.ca/faune/especes/menacees/liste.asp
158
At least one, at least a few, a few, identity, most, almost all, all (Malczewski,
2006a).
At least one, at least a few, a few, identity, most, almost all, all (Boroushaki
and Malczewski, 2008).
Air quality : very clean – severely polluted (0 – 125)( Sumathi et al., 2007).
Dans notre cas, ces échelles ont représenté un repère pour la construction des échelles
ordinales d’indicateurs non quantitatifs. Il a été décidé de ne pas utiliser un type nine-point
scale du fait qu’un large éventail de possibilités exigeait plus de temps de discussion aux
acteurs, le temps étant une contrainte pour cette recherche. De plus, un nombre plus restreint
d’échelons permet un choix net entre les options proposées. Ainsi, seulement quatre échelons
ont été choisis. Le nombre paire d’échelons évite l’utilisation d’une position moyenne et
demande un choix clair.
Après cette revue de littérature, la construction d’échelons a été réalisée par une
analyse spatiale complétée, lorsque nécessaire, par des calculs avec un classeur Excel.
performances de chaque scénario sur chaque critère, au moyen des échelles de mesure
associées.
Le « type » fait référence aux deux possibilités d’évaluation que le logiciel D-Sight
offre aux usagers25 : soit une comparaison par paire, Pair Wise, ou par utilité, Utility. Pair
Wise fait référence à la méthode retenue pour ce test, PROMÉTHÉE (voir Vazquez, Waaub
et Ilinca, 2013b) tandis que Utility fait référence à une autre méthode d’évaluation : Multi-
Attribute Utility Theory26 (MAUT).
La fonction de préférence est en lien avec le critère. Le critère, « c’est une référence
par rapport à laquelle on mesure les conséquences d’une action » (Schärling, 1985,
page 182). La méthodologie choisie, PROMÉTHÉE, propose six types de critères dont deux
font partie de ceux retenus pour ce test. Ils sont décrits ci-dessous.
Le vrai critère « (sans paramètre) » (Schärling, 1985, page 184). Appelé aussi
« Type 1 : Usual Criterion » (Brans et Mareschal, 2005, page 170) ou critère « franc »
(Schärling, A., 1996, page 134) est sans paramètre à fixer. Selon ce type de critère, toute
différence de performance entre deux scénarios pour un même critère entraîne
automatiquement une préférence stricte (préférence ayant la valeur de 1) pour celui qui a la
meilleure valeur. Il y a indifférence (préférence ayant la valeur de 0) lorsque les
performances sont égales.
25
http://www.d-sight.com/manual/manual.html Page Web consultée le 2 mars 2013
26
http://www.d-sight.com/manual/manual.html Page Web consultée le 2 mars 2013.
160
Le premier critère, Usual, sera la fonction utilisée pour les échelles qualitatives,
ordinales. C’est une fonction qui établit la préférence complète d’un scénario sur un autre dès
qu’une différence de performance apparaît (ex. : appartenance à une classe). Le deuxième
critère, V-shape, sera la fonction utilisée pour les échelles quantitatives, cardinales. C’est une
fonction qui discrimine un scénario par rapport à un autre sur la base de la moindre différence
d’écarts.
Pour ce qui est de la valeur absolue ou relative, le logiciel D-Sight permet de choisir
entre des seuils basés sur l’échelle d’évaluation (absolue) ou sur des seuils en pourcentages
(relatifs), respectivement. Du fait que ce test considère une échelle de mesure, l’évaluation est
faite en mode absolu.
Finalement, la pondération des critères est différente pour chaque acteur et constitue
ses préférences particulières, ses priorités en fonction de son système de valeurs personnel ou
organisationnel quand il représente une organisation.
Cette annexe présente un complément des résultats du test de DEMIT (objectif 2) non
précisés dans l’article à publier (chapitre II) en raison d’un manque d’espace.
À partir d’une feuille de travail, les acteurs participants ont choisi un poids de pouvoir
décisionnel égal pour chacun, peu importe la catégorie (experts, société civile, secteur privé
ou secteur public) à laquelle ils appartenaient. Chaque acteur a reçu donc un pourcentage égal
à 9,09 %. Ce pourcentage définit le Modèle numéro 1. Ainsi, la catégorie ayant le plus grand
nombre de membres – la société civile – en bénéficie davantage. Les résultats détaillés du
Modèle 1 ont été présentés dans Vazquez, Waaub et Ilinca, 2013b. Par ailleurs, en raison des
limites de cette recherche, la discussion sur l’analyse de préférence, la consolidation du
jugement – incluant l’analyse de sensibilité et l’analyse de robustesse – et la décision
concerne uniquement ce Modèle 1 et elle a été entièrement réalisée dans l’article indiqué.
Par ailleurs, trois choix de poids de pouvoir décisionnel sont suggérés (voir Vazquez,
Waaub et Ilinca, 2013b) et, par conséquent, trois modèles. Les résultats en ce qui concerne la
classification des scénarios et des coalitions des acteurs (Plan GAIA) des Modèles 2 et 3 sont
présentés dans cette même annexe (voir figures B.1 et B.2). Le Modèle 2 attribue 25 % du
pouvoir décisionnel à chaque catégorie des acteurs (privé, public, société civile et experts).
Dans le Modèle 3, 50 % du pouvoir décisionnel appartient au secteur privé, 25 % au secteur
public, 12,5 % à la société civile et le 12,5 % restant aux experts.
164
1. Accidents de circulation.
2. Accomplir les objectifs de production (optimiser la configuration : non pas
toutes les superficies possèdent le même potentiel éolien; par exemple, sur
65 km2 le potentiel n’est pas distribué uniformément).
3. Amélioration de l’accès routier, réaménagement des chemins, dû aux travaux
réalisés lors de la construction du parc.
4. Aménagement des paysages plutôt que protection des paysages et en
conformité avec les attentes des communautés.
5. Apparition de problèmes d’incompatibilité d’usage. En 2002, lors du premier
appel d’offres de 1000 Mégawatts d’énergie éolienne, la ville de Gaspé ne
comptait pas avec une réglementation pour encadrer l’usage éolien du
territoire.
6. Appel d’offres pour les entreprises de la région (expert-conseil).
7. Banalisation du paysage (alignement des éoliennes, lignes électriques =
structures industrielles similaires).
8. Baisse du prix des maisons (en Europe : les habitations situées à 500 m d’une
éolienne sont invendables).
9. Bonne productivité des gisements éoliens.
10. Compensation écologique (revitalisation des milieux dégradés).
11. Compensations justes pour tous – incluant l’impact visuel sur les citoyens qui
ne reçoivent pas de redevances.
12. Conflits d’usage. 1) Agriculture-éolien : clause interdisant la construction
d’un silo permettant d’alimenter des animaux près d’une éolienne;
construction d’usines sur des terres arables. 2) Tourisme-éolien : manque
d'un plan pour protéger l'industrie touristique – de haute importance
économique – qui repose sur la qualité des paysages; aperçu sommaire des
impacts potentiels des parcs éoliens sur les paysages de la Gaspésie et de
l'industrie touristique de la Gaspésie-Îles-de-la-Madeleine. Dans la
période 2002-2010, les dépenses touristiques ont été de 1,9 milliard de
dollars. Pour sa part, les investissements totaux de 1000 MW éoliens (2006-
2012) sont calculés à 2,1 milliards de dollars (colloque AQPER). Il n'existe
pas d'étude d'impact visuel sur le tourisme maritime.
13. Conflits d’utilisation du territoire. (Relocalisation des résidences déjà établies
versus relocalisation des éoliennes.) (Industrie éolienne versus industrie
touristique.)
14. Connaître les besoins réels en énergie.
15. Connaissance des besoins réels en énergie. Consultation avant l’application
du développement de la filière éolienne.
16. Consultation à un niveau de planification stratégique, pas après les faits.
17. Consultation auprès des impliqués.
18. Contrats : 20 ans, pouvant être renouvelés pour aller jusqu’à 50 ans.
165
19. Crédits de carbone (cycle complet du carbone lors de toutes les étapes :
construction, opération, démantèlement).
20. Charges financières à long terme pour les municipalités : démantèlement,
enfouissement sanitaire, détérioration des routes, entretien des chemins aux
frais des contribuables.
21. Chasse : l’étude d’impact ne s’est pas intéressée ni aux orignaux ni aux
chevreuils.
22. Choix d’un partenaire privé reconnu et impliqué dans l’exploitation des parcs
éoliens.
23. Db à la distance de 200 m du sentier international des Appalaches.
24. Des lois pour exiger et non pas pour faire des recommandations ou des
suggestions au promoteur du projet = approbation du projet en vertu du plan
d’implantation et d’intégration architecturale (PIA). Respect des lois et
règlements existants (SDA, PIA, RCI éolien, RCI forêt, protection des rives,
du littoral et des plaines inondables des lacs et cours d’eau, PPMV, plan
d’urbanisme, plan d’implantation et d’intégration architecturale [articles 5 et
6 de l’objectif 2 : fils souterrains et aménagement paysager]).
25. Déversements de produits, d’huile (altération de l’eau potable).
26. Développement éolien selon les droits – ignorés – de la nation Mi’gmaq :
demande de consultation selon le protocole des Premières nations (le Plan
régional de développement du territoire public, volet éolien, ne l’a pas pris en
compte) et demande d’adaptation des consultations du BAPE au protocole de
consultation des Premières nations.
27. Développement de l’éolien sans bien connaître les besoins réels en énergie.
28. Développement d’une main-d’œuvre québécoise qualifiée dans le secteur
éolien.
29. Prise en compte d’une distance séparatrice entre les résidences et les
éoliennes.
30. Diversification des activités économiques (tourisme plus « éolien ») :
création d’usines (assemblage et fabrication de composantes).
31. Effet stroboscopique (lumière sur les pales = nausées, épilepsie chez les
personnes fragiles), donc distance séparatrice de plus de 2 km.
32. Effet sur le tissu social (après projet / balance de pouvoir / propriétaires de
terrain).
33. Élaboration d’un cadre d’implantation basé sur la protection et
l’aménagement des paysages et non sur l’exploitation de l’énergie éolienne.
34. Élaboration d’un Plan de développement éolien global. Plan d’ensemble et
des normes précises.
35. Élaboration d’un plan de paysages à conserver et mise en marche.
36. Emplois : De quels types? Dans quelles conditions salariales?
37. Éoliennes autorisées à une distance de 200 mètres du sentier international des
Appalaches.
38. Éoliennes, plus de monuments que de bâtiments.
39. Équité de traitement pour tous les propriétaires fonciers.
166
114. Règles à établir autant sur les terres publiques que sur les terres privées.
115. Réinvestissement en faveur de la faune : inventaires des oiseaux.
116. Relocalisation des éoliennes (non au sommet des pentes).
117. Respect de la volonté des propriétaires de terrains tout en rencontrant les
objectifs de production de l’entreprise (configuration du parc à partir des
préférences des propriétaires).
118. Respect de la qualité des paysages en bordure de la route 132 (corridor
touristique de la route 132). Aucune éolienne entre la mer ou le Saint-Laurent
et la route 132.
119. Respect des lois et normes environnementales.
120. Respect de l’optique de la Politique du Québec sur le tourisme et le
développement durable.
121. Respect des produits touristiques de la Gaspésie centrés sur l’observation du
paysage créateur d’émotion chez les touristes.
122. Respect de toutes les lois applicables au secteur éolien.
123. Respect du budget d’aménagement.
124. Retombées économiques pendant 20 années (durée de vie du projet).
125. Risque de détérioration de la qualité des sols et des eaux (déversement
d’hydrocarbures et fuites d’huile, érosion associée au réseau routier,
exploitation de gravier, construction en béton). Déforestation (accès, aires de
travail, lignes de transport d’énergie. Cimetières d’éoliennes (art. 1116 du
Code Civil du Québec : enlever les infrastructures… Sauf si lesdites
infrastructures ont plus de valeur que le terrain… = la compagnie peut
acheter le terrain pour la valeur d’évaluation et y laisser ses infrastructures).
126. Risque potentiel pour la faune (terrestre, aviaire, aquatique). Les animaux
sont plus sensibles que l’être humain aux infrasons. Faune ailée : avifaune et
chiroptères (risque de mortalité par collision et par perte d’habitat).
127. Sessions hiver / été (conditions différentes, différents impacts).
128. Suivi sur la migration des oiseaux. Suivi sur la mortalité des oiseaux.
129. Trafic lourd, vitesse excessive, poussières.
130. Traitement équitable envers toute personne, d’un projet à l’autre
(rémunération : tous sur la même base).
131. Un site patrimonial de grande valeur (phare de Pointe-à-la-Renommé) est un
produit touristique structurant pour l’économie de la région.
132. Valeur historique du paysage culturel.
133. Versements financiers aux municipalités : contribution volontaire annuelle,
fonds de visibilité de 30 000 $ aux municipalités, accord d’équité qui prévoit
les mêmes avantages si des contributions plus avantageuses sont allouées
ailleurs.
134. Voies d’accès existantes à proximité du parc.
135. Atteinte d’objectifs du protocole de Kyoto*.
136. Lutte contre les changements climatiques ((cycle de vie – bilan énergétique
de l’éolien).*
137. Quantité de CO2 évitée*.
170
Ensuite, les préoccupations ont été structurées par enjeu, dont 86 appartiennent à la
dimension socioculturelle, 34 à la dimension économique et 17 à la dimension
environnementale.
De l’avis des acteurs, la première des cinq a été réservée pour une préoccupation en
lien avec la perception de l’acceptabilité sociale. La deuxième des cinq a été réservée pour
une préoccupation concernant l’impact économique dans la communauté hôte. Les trois
dernières – en bleu – ont été classifiées dans la dimension environnementale. Cependant, à la
suite d'une discussion, la table de dialogue et négociation a décidé de ne pas les inclure dans
le choix de critères et de considérer seulement 14 préoccupations environnementales. Pour
prendre cette décision, une recherche d’informations sur le bilan énergétique de l’éolien a été
réalisée dans le Web.
Ledit bilan (énergétique incluant les émissions de CO2) est positif27 du fait que, pour
une éolienne moderne, l’énergie nécessaire à sa fabrication, son transport, son installation, sa
maintenance et son démantèlement sont récupérés en de 2 à 3 mois de fonctionnement 28.
Étant sa vie moyenne de 20 ans, le bilan énergétique de l’éolien s’avère écologique.
27
http://www.windpower.org/en/. Site Web consulté le 17 juin 2010.
28
http://www.notre-planete.info/ecologie/energie/energie_eolienne_0.php. Site Web consulté le 20 mars
2012.
171
Par ailleurs, trois critères ont été inspirés du projet Suivi des indicateurs de
l’aménagement durable des forêts du ministère des Ressources naturelles et de la Faune du
Québec (projet INDI)29, ce qui a permis de réaliser cette étape plus rapidement que prévu du
fait de compter avec des critères déjà structurés et éprouvés. Ces critères, identifiés en mauve
sont :
Ensuite, ces huit critères ont été précisés davantage, pour mieux représenter les
préférences des acteurs. Ainsi, l’impact sur la santé a été divisé en deux sous-critères,
l’amélioration de la qualité du paysage en trois et l’impact économique dans la communauté
hôte en trois. Cela a donné lieu à la proposition finale, qui inclut un total de 13 critères.
29
http://www.mrn.gouv.qc.ca/publications/enligne/forets/criteres-indicateurs/accueil.asp. Page Web
consultée le 17 juin 2010
172
Tableau B.1 Critères proposés par DEMIT, classifiés selon les dimensions du
développement durable.
SOUS-CRITÈRE
DIMENSION CRITÈRE
(devient un critère)
1. Exposition des résidences dans
1. Impact sur la santé et la le domaine du parc.
sécurité.
2. Diminution potentielle de
l’espérance de vie.
2. Amélioration de la 3. Lieux significatifs.
qualité du paysage.
4. Appartenance au territoire.
(Collaboration : Marie
Sociale
dans la communauté
tourisme.
hôte.
10. Emplois.
6. Avantages socio- 11. Avantages socio-économiques
économiques régionaux régionaux du projet.
du projet.
7. Dégradation du milieu 12. Dégradation du milieu
Environnementale physique. physique.
8. Perte de la biodiversité. 13. Perte de la biodiversité.
Dans ces critères, l’exposition des résidences dans le domaine du parc fait référence
aux préoccupations concernant les résidences se trouvant à 500 m ou moins des éoliennes. La
diminution potentielle de l’espérance de vie fait référence aux préoccupations sur les
résidences à 750 m des éoliennes. Cela pour illustrer deux cas de distances séparatrices. La
première indiquée dans le Règlement de contrôle intérimaire (RCI) de la MRC de Matane –
et à l’intérieur du domaine du parc – et la deuxième, supérieure à 500 m et hors du domaine
du parc éolien. Ce critère met en évidence qu’il existe de situations tant à l’intérieur qu’à
173
Échelles de mesure.
Les échelles de mesure ont été présentées à la table de dialogue et négociation pour
leur validation. Les échelles de mesure de chaque indicateur sont présentées ici.
Critère : Exposition des résidences dans le domaine du parc, selon les distances
séparatrices.
Cette distance a été sélectionnée en tenant compte du RCI de Matane (RCI 220-3-
2007) qui, à la page 4, dit : « En dehors des périmètres d’urbanisation, toute éolienne doit être
située à au moins 500 mètres de toute résidence et aucune nouvelle résidence ne peut être
implantée dans cette bande de protection de 500 mètres ».
Des 7 bâtiments trouvés, pour valoriser les écarts, un facteur égal à 22,57 a été utilisé
pour associer l’échelle au nombre total de bâtiments qui se trouvent dans le domaine du parc,
soit 158 (7*22,57= 158). Il s’agit d’un facteur symbolique concernant le total des bâtiments
dans le domaine du parc.
Il est important de souligner que le parc éolien qui fait l’objet du test de l’approche
DEMIT –Baie-des-Sables – a été mis en opération en novembre 200630 et que, dans son étude
d’impact sur l’environnement, à la page 3-2, on peut lire : « L’emplacement des turbines
respecte les distances minimales suivantes :… 500 mètres de toute résidence » (Cartier
énergie éolienne, Pesca Environnement et Hélimax Énergie, 2004).
La contradiction entre ce qui est dit dans l’étude d’impact (0) et les sept bâtiments
trouvés dans la couche provenant de la carte topographique 22B12-0201 pourrait venir d’une
actualisation de l’information de 2005 à l’année de l’achat de la carte – 2010 – ou du fait que
les bâtiments indiqués dans la carte 22B12-0201 ne soient pas des résidences. Cependant,
pour ces 7 bâtiments trouvés, ladite carte indique – dans la description – « Bâtiment ». En
raison des limites de cette recherche, aller sur le terrain pour les vérifier n’était pas une option
envisageable. En conséquence, les sept bâtiments identifiés dans la requête spatiale ont été
considérés.
30
http://www.cartierenergie.com/cartier-energie.aspx?sec=7, page d’internet consultée le 18 mars 2013.
175
Ainsi, la division de l’échelle de cet indicateur, qui varie de 0 à 158, est définie comme
suit et avec un écart quasi proportionnel de 50 unités :
ÉCHELLE
Aucun : 0
Bas : 50
Moyen : 100
Haut : 158
Il s’agit donc d’une échelle quantitative sur la base exclusive d’une analyse spatiale.
Pour construire cette échelle, toutes les résidences dans un rayon de 750 m des
éoliennes – peu importe qu’elles soient incluses ou non dans le domaine du parc – sont
considérées. Cette distance, arbitraire dû aux limites de cette recherche, illustre le cas d’une
distance supérieure au RCI de Matane et reflète les préoccupations exprimées dans les
mémoires des audiences publiques sur ce parc. D’après la requête spatiale faite, un total de
176
150 bâtiments est concerné pour le scénario 1 (voir annexe D4). Le total pour les scénarios 2
et 3 est de 78 bâtiments chacun (résultat non illustré sur une carte du fait que c’est la même
procédure que pour le scénario 1). La division de l’échelle varie donc entre 0 et 150 unités,
avec un écart quasi proportionnel de 38 unités.
ÉCHELLE
Aucun : 0
Bas : 40
Moyen : 78
Haut : 150
Ainsi, il s’agit d’une échelle quantitative sur la base exclusive d’une analyse spatiale.
ÉCHELLE.
1. Non
Les échelons de 1 à 4, en nombre réduit et pair, ont été sélectionnés afin d’obliger à
prendre nettement une position. D’abord, il s’agit d’une décision oui/non pour ensuite noter
la force du oui.
ÉCHELLE
1. Non
2. Bas
3. Moyen
4. Haut
178
ÉCHELLE
1. Non
2. Bas
3. Moyen
4. Haut
Il a été observé que, entre 1997 et 2010, les conclusions des rapports du BAPE au sujet
des projets éoliens portaient, en général, sur la façon de dissiper les incertitudes concernant
des enjeux majeurs. En 2006, un projet a été effectué en contravention de la réglementation
en vigueur de la MRC correspondante. En 2010, il est indiqué que des projets sont
acceptables sur le plan biophysique et économique. Par contre, une problématique
particulière a été identifiée sur le plan social, et on peut y rencontrer les expressions : impact
psychosocial, détérioration des relations entre partisans et opposants du projet, réalisation de
référendum pour connaître la position réelle de la population sur le projet. Ainsi, les
expressions retrouvées dans ces conclusions au sujet des inconforts de la population ont été
prises en compte pour construire cette échelle.
ÉCHELLE (d’après les conclusions des rapports des projets éoliens du BAPE)
3. Population inquiète
4. Aucune tension
Ainsi, pour mesurer ce critère 6, une autre échelle qualitative a été créée – transposée à
des valeurs numériques – représentant quatre différentes manières dont la construction d’un
180
ÉCHELLE
1. Aucune : 159
2. Bas : 108
3. Moyen : 84
4. Haut : 0
Il s’agit donc d’une échelle quantitative sur la base d’une analyse spatiale et d’un
classeur Excel pour comptabiliser les superficies (voir annexe D5 pour la carte et annexe E1
pour les calculs). Les résultats des scénarios 2 et 3 ne sont pas illustrés sur une carte du fait
que c’est la même procédure que pour le scénario 1.
ÉCHELLE
Aucun : 0
Bas : 87
Moyen : 110
Haut : 144
Ainsi, il s’agit d’une échelle quantitative sur la base d’une analyse de coûts par année.
La base de ce coût est de 1000 $ par mégawatt auquel s’ajoute un fond de visibilité de 30
000 $ par année (voir annexe E2). Cette base illustre une façon de procéder au calcul des
avantages financiers potentiels, mais d’autres bases auraient pu être prises en compte en
incluant des montants spécifiques pour les loyers aux propriétaires des terrains.
Tableau B.6 Construction des échelons pour évaluer le critère 9 « impact négatif
potentiel sur le tourisme »
ÉCHELLE
Aucun : 0
Bas : 90
Moyen : 175
Haut : 329
Il s’agit d’une échelle quantitative sur la base d’une analyse spatiale et d’un autre sur
classeur Excel, tout en considérant des catégories de paysage.
Critère : Emplois
ÉCHELLE
Aucun : 0
Bas : 78
Moyen : 106
Haut : 160
Il s’agit d’une échelle quantitative sur la base d’une analyse sur classeur Excel, en
considérant le nombre moyen de travailleurs en période de construction et le nombre
d’employés en période d’exploitation (voir annexe E4).
ÉCHELLE
Aucun : 0
Bas : 51,2
Moyen : 98,4
Haut : 102,4
Il s’agit d’une échelle quantitative sur la base d’une analyse sur classeur Excel, en
considérant un pourcentage de l’investissement total en dépenses dans la région l’accueillant
(voir annexe E5).
ÉCHELLE (m2x103)
Aucun : 0
Bas : 20
Moyen : 40
Haut : 80
186
Il s’agit d’une échelle quantitative sur la base d’une analyse spatiale pour repérer le
nombre d’éoliennes dans la catégorie Sol sensible et d’un classeur Excel pour comptabiliser
les superficies (voir annexe C6 pour la carte et annexe E6 pour les calculs).
ÉCHELLE (x 103)
Aucun : 0
Bas : 110
Moyen : 220
Haut : 460
Il s’agit d’une échelle quantitative sur la base d’une analyse spatiale pour repérer le
nombre d’éoliennes dans les zones de cédrières, d’érablières, de forêts de feuillus mixtes et
de forêts de résineux mixtes et d’un classeur Excel pour comptabiliser les superficies (voir
annexe D7 pour la carte et annexe E7 pour les calculs).
187
Impact négatif potentiel sur le Pair Wise Minimiser V-Shape Absolue 329
tourisme
Quant aux résultats sur les 11 acteurs participants, ils sont répartis dans les quatre
catégories de la manière suivante :
La participation de chaque acteur dans chaque étape de DEMIT a déjà été abordée dans
Vazquez, Waaub et Ilinca, 2013b.
Classification des scénarios et Plan GAIA (Global Visual Analysis) des modèles 2 et 3.
Les résultats du modèle 1 ont été détaillés dans Vazquez, Waaub et Ilinca, 2013b.
Ici, nous présentons les résultats concernant la classification et le plan GAIA des
modèles 2 et 3. Les quatre scénarios à classer sont :
Scénario 4 : 0 turbine
191
0.3
Classification des scénarios: Modèle 2
0.2
0.21
0.16
0.1
0 -0.03
-0.1
-0.2
Poids de pouvoir décisionnel de 25% pour
-0.3 -0.35 chacune des quatre catégories d'acteur
-0.4
Scénario 1 Scénario 2 Scénario 3 Scénario 4
Nous observons, dans la figure 3.1 : à égale répartition du pouvoir décisionnel entre les
acteurs, le scénario 3 est préféré (construction du parc en changeant la technologie pour
réduire le nombre d’éoliennes tout en respectant les objectifs de production de l’investisseur
du projet). Bien que le scénario 4 (non-construction du parc et, donc, aucune éolienne à
implanter) soit aussi préféré, une négociation s’avère un outil pour concrétiser la construction
du scénario 2 tout en intégrant les préoccupations représentées dans la note de 0,16 octroyée
au scénario 4.
0.3 192
Classification des scénarios: Modèle 3
0.25
0.27
0.2
0.15
0.1
0.05
0 -0.05 -0.06
-0.05
-0.1
-0.17
-0.15
-0.2
Scénario 1 Scénario 2 Scénario 3 Scénario 4
Nous observons dans la figure B.2 que, pour un pouvoir de décision favorisant
davantage l’investisseur du projet, un scénario permettant la construction du parc est
largement préféré. Par rapport au modèle 2, la note du scénario de non-construction est
réduite de 22 unités.
Note
Figure B.3 Plan GAIA du modèle 2 : position de l’axe de décision pour un pouvoir
décisionnel de 25 % par catégorie d’acteur. Interprétée de View Global Results
194
Figure B.4 Plan GAIA du modèle 3 : position de l’axe de décision pour un pouvoir
décisionnel de 40 % pour le promoteur du projet. Interprétée de View Global Results
La distribution des alternatives sur les deux dimensions montre que les scénarios 1 et
2, qui se trouvent dans le même quadrant, ont certaines ressemblances. Par contre, leur
éloignement dans le cas des scénarios 3 et 4 indique une différence significative. La longueur
de l’axe de décision, large, montre une décision solide.
ANNEXE C
Code couleur :
Vert : préoccupations exprimées dans les mémoires déposés auprès du BAPE, en 2005.
Rouge : préoccupations exprimées dans les mémoires déposés auprès du BAPE, en 2005.
Bleu : préoccupations exprimées dans le test de DEMIT, dans la période 2010 - 2011.
Orange : enjeux du promoteur exprimés lors de la réalisation du test, en 2010 et 2011.
Mauve : critères inspirés du projet INDI du MRN-Québec, pour l’aménagement durable des
forêts.
196
3 Effet stroboscopique (lumière sur les pales = nausées, épilepsie chez les personnes fragiles), donc
distance séparatrice de plus de 2 km.
4 Infrasons (bruit sourd rapproché du rythme cardiaque = irrégularité du pouls = états anxieux).
5 Incendies.
6 Pollution auditive /bruit des éoliennes/ niveau d’augmentation du bruit (de 40 db à une distance
de 500 m) avec l’usure et le temps.
9 Impacts électromagnétiques.
197
Tableau C.3 Regroupement des préoccupations sur la prise en compte des valeurs
et des besoins des populations
2 Faisabilité du projet.
ENJEU
3 Chasse : l’étude d’impact ne s’est pas intéressée ni aux orignaux ni aux chevreuils.
7 Risque potentiel pour la faune (terrestre, aviaire, aquatique). Les animaux sont plus
sensibles que l’être humain aux infrasons. Faune ailée : avifaune et chiroptères (risque de
mortalité par collision et par perte d’habitat).
8 Suivi sur la migration des oiseaux. Suivi sur la mortalité des oiseaux.
207
2 Amélioration de la qualité du
paysage (maximiser)
9 14 3 3 29
3 Prise en compte des valeurs et
des besoins des populations
(maximiser)
14 13 1 28
4 Perception de l’acceptabilité
sociale (minimiser)
TOTAL 86
12 12 2 26
5 Impact économique dans la
ÉCONOMIQUE
communauté hôte
(maximiser)
6 2 8
6 Avantages socio-économiques
régionaux du projet
(maximiser)
TOTAL 34
2 3 1 6
7
ENVIRONNEMENTAL
Dégradation du milieu
physique (minimiser)
2 6 8
8
Perte de biodiversité
(minimiser)
TOTAL 14
TOTAL 52 69 6 7 134
ANNEXE D
CARTES THÉMATIQUES
209
Figure D.3 Différenciation entre les scénarios 1 et 2 dans le rayon de 500 mètres
212
Figure D.4 Identification des 150 bâtiments dans la zone tampon de 750 mètres
213
Figure D.7 Éoliennes identifiées dans les cédrières, érablières, peuplements feuilllus
et résineux mixtes
ANNEXE E
Superficie
Construction Électrification Emplacement Nombre Capacité F1* F2** Superficie Superficie
multiplié ha.
de routes souterraines de turbines de MW (%) m2 par
F2
Bande Longueur Bande Longueur m2 turbines
Scénario 1 10 11 629 8 18 300 20 000 32 1,5 0 76 902 690 1 588 734,4 159
Scénario 4 10 0 8 0 20 000 0 0 0 0 0 0 0
*F1 : Facteur 1
Facteur, arbitraire en raison des limites de cette recherche, sur la bande d'électrification
souterraine.
Pour indiquer un changement dû aux plus grandes dimensions des éoliennes de 3 MW.
**F2 : Facteur 2
Facteur, arbitraire en raison des limites de cette recherche, pour indiquer une
perception différente de l'affectation à la culture, par exemple, sur la variation du microclimat
ou sur une plus grande dimension des pales des éoliennes de 3 MW.
Montant de base de 1 000 $ par mégawatt plus un fond de visibilité de 30 000 $ par année.
Cette base illustre une façon de procéder au calcul des avantages financiers potentiels, mais
d'autres bases auraient pu être prises en compte et inclure des montants spécifiques pour les loyers
versés aux propriétaires des terrains.
219
Scénario 1 55 18 5 3 329
Scénario 2 29 9 5 3 172
Scénario 3 29 9 5 3 172
Scénario 4 0 0 5 3 0
Facteur de pondération :
Attribué selon les catégories de paysage
Ce facteur sert à écarter les différences davantage
Valeurs de 1, 2 et 3. Choix arbitraire, selon l'exemple de la fiche d'inventaire Ruralys et
al., 2008, page 131.
La somme a été ajustée : de 172 à 175.
220
Nombre de MW Phases
turbines turbines Construction Opération Total
Scénario 1 73 1,5 150 10 160
Scénario 2 38 1,5 73 5 78
Scénario 3 38 3 101 5 106
Scénario 4 0 0 0 0 0
Les scénarios 2 et 3 ont reçu une valeur – arbitraire en raison des limites de cette
recherche – représentant une variation selon le nombre et les caractéristiques des turbines
considérées.
Ces valeurs n’excèdent pas les chiffres indiqués dans la page Internet consultée*
Investissement* % dépensé**
Total
M$ dans la région
Scénario 1 164 0,6 98,4
Scénario 2 85,3 0,6 51,18
Scénario 3 170,7 0,6 102,42
Scénario 4 0 0,6 0
** 60 % de l’investissement total est considéré comme adéquat pour être dépensé dans
la région. Ce pourcentage est mentionné dans différents documents, notamment celui de la
Direction des évaluations environnementales, août 2011.
Emplacement Nombre d’
éoliennes dans Total m2x103
de turbines (m2) sol sensible
Scénario 1 20 000 4 80 000 80
Scénario 2 20 000 0 0 0
Scénario 3 20 000 0 0 0
Scénario 4 20 000 0 0 0
La superficie calculée, pour les terres agricoles, n’excède pas celle indiquée dans
l’étude d'impact sur l'environnement (page 5-37) et est égale à 95 ha, Cartier énergie
éolienne, Pesca Environnement et Hélimax Énergie, 2004.
223
Flore Faune
Emplacement Nombre d’éoliennes
des turbines dans : Nombre d’éoliennes dans : TOTAL
m2x103
(m2)
(m2) Feuillus Résineux
Cédrière Érablière
mixtes mixtes
Scénario 1 20 000 3 3 13 4 460 000 460
Scénario 2 20 000 2 0 7 2 220 000 220
Scénario 3 20 000 2 0 7 2 220 000 220
Scénario 4 20 000 0 0 0 0 0 0
Les feuillus mixtes et les résineux mixtes ont été choisis comme habitat représentatif de la
faune, parce que « les feuillus mixtes représentent la strate la plus répandue dans le secteur à l’étude »
et qu'ils « couvrent la plus grande superficie forestière du site du parc éolien, et ce, pour la classe d’âge
de 50 ans ». (Cartier énergie éolienne, Pesca Environnement et Hélimax Énergie, 2004. Page 2-14)
(étude d’impact). Les résineux mixtes, pour ces mêmes caractéristiques, les suivent en importance
(tableau 2.5 Répartition des strates forestières [hectares] par classe d'âge dans la même étude, p. 2-
13).
De plus, il est considéré que « les feuillus accueillent plus de diversité, car ils secrètent peu de
composés toxiques ou acides, mais les résineux accueillent des espèces qui leur sont très spécifiques »,
Lair, P., 2011, page 5.
En outre, il est considéré qu’au Québec « la zone de la forêt feuillue possède une très grande
valeur du point de vue de la biodiversité »
http://www.scienceenjeu.com/forestia/fiche/fiche_feuillue.pdf. Page consultée le 26 mars 2013.
L’érablière a été choisie parce que dans l'étude d'impact du projet (page 2-13) indique qu’elle
fait partie de la vieille forêt inéquienne. La cédrière a été choisie parce qu'elle est le seul type de
peuplement forestier rapporté dans l'étude d'impact (page 2-13) dans la tranche de 50 à 120 ans.
En raison des limites de cette recherche, aucune autre information spatialisée, à part le boisé, ne
pouvait être utilisée pour construire ces échelons. Ainsi, cédrière et érablière ont été choisies pour
représenter la flore touchée, et résineux et feuillus mixtes l’ont été pour représenter l'habitat nécessaire
à la faune.
La superficie calculée sur les terrains de ces forêts n'excède pas celle indiquée dans l'étude
d'impact sur l'environnement (page 5-37) (Cartier énergie éolienne, Pesca Environnement et Hélimax
Énergie, 2004) et égale à 107 ha.
ANNEXE F
FACTEURS
Choix d’information géoréférencée
Faire ressortir Possibilité de participation Prise en compte des Réalisation d’une étape du
l’information existante multi-acteurs. critères que chaque acteur module SIG et préparations
dans son contexte spatial. Partage des idées, discussion. défend. d’autres.
Validation de l’information et Participation de tous les Établissement de règles de
de la qualité de la donnée. acteurs (SIG participatifs et prise de parole.
Possibilité de prise en collaboratifs).
FORCES
considération de bases de
données de création locale.
(1) (4) (2) (2)
INTERNE
(1)
TOTAL 4 8 6 4
FACTEURS
Support informatique (ordinateur et logiciel)
Capacité de stockage de Accès facile à l’information Facilite l’échange d’idées Réalisation d’une
données.Édition de l’information. stockée. (d’appui au processus cognitif étape du processus
Performance des Partage de données. des acteurs) sur les critères à de décision et
outils du logiciel. Accès et copie de l’information analyser. préparations à des
FORCES
Coûts associés (i.e. équipe, logiciel Risque de tendance à la Perception des acteurs sur
FAIBLESSES
TOTAL 12 8 4 4
FACTEURS
Obtention de données (numérisation et carte numérique)
Possibilité d’obtenir information Augmenter la quantité Accès à information Amélioration d’une étape
vectorielle à partir d’un document non d’information visuelle pour un échange du module SIG grâce à
disponible en format numérique. (numérique) disponible. d’idées plus facile. l’augmentation
Si existence de produits numériques, accès Accès rapide à Appropriation et d’information numérique.
à information vectorielle facilement l’information visuelle identification du Réalisation d’une étape du
FORCES
reproductible et qui permet la sur différents éléments territoire grâce à processus de décision et
compréhension du territoire. du territoire. l’identification des préparation à d’autres.
éléments importants
pour les acteurs.
INTERNE
salaires. données.
Incertitude de données spatiales Jeux politiques.
(conception, mesurage, représentation et Risque de privilégier
analyse). l’information déjà en
format numérique plutôt
que la plus pertinente
(2) (3)
228
FACTEURS
Obtention de données (numérisation et carte numérique)
données numériques.
Échelle cartographique et/ou
résolution.
Usagers multiples et concurrents
(4)
(1)
TOTAL 12 8 2 2
FACTEURS
Conception des scénarios
Conception participative des Information visuelle pour alimenter Intégration des valeurs de Réalisation d’une
scénarios à être analysés. la discussion sur différents éléments chaque acteur, i.e. étape du processus de
Élaboration de zones du territoire et sur la distribution des identification des éléments décision et
d’interdiction selon différents éoliennes. sensibles du territoire du préparations à
FORCES
R&D Règles flexibles dans les appels R&D Accès facile aux RCI
OPPORTUNITÉS
(en SIG participatifs et d’offres de projets éoliens (i.e. (en SIG participatifs et (cadre réglementaire :
collaboratifs). permettant un changement sur la collaboratifs). distance séparatrice
technologie choisie et donc un des éoliennes)
scénario différent)
EXTERNE
RCI (cadre
réglementaire)
(1)
TOTAL 6 3 3 3
FACTEURS
Conception des cartes
possible. (1)
R&D. (4)
TOTAL 9 3 4 2
BESOINS/ 1. Politisation
DÉFIS 2. Exigence du temps alloué à l’analyse de la carte.
3. Experts en SIG participatifs et collaboratifs.
4. Conception SIG-WEB (cartographie Web) pour diffusion
231
MODULE AMCD
Possibilité de ralentir le
processus dû à un manque de
langage commun ou à un
manque de disponibilité des
acteurs.
(1)
232
MODULE AMCD
Manque d’expérience de l’utilisation des D’anciens conflits vécus Absence d’une structure
outils de décision. dans la communauté qui favorisant un espace
EXTERNE
décision.
Ne pas s’impliquer à cause de l’inconnu :
la maitrise de l’outil échappe aux acteurs
sociaux.
Enjeux qui ne son pas traites par les
acteurs (i.e. dans les mémoires du BAPE).
Un résultat du processus AMCD qui ne
soit pas représentatif de la réalité.
TOTAL 1 18 4 4
BESOINS/ 1. Lors des débats, surmonter des confrontations possibles entre les acteurs, dues à des espaces de rapport
DÉFIS de forces.
2. Conjoncture démocratique.
3. Appuis financiers (Coûts associés).
4. Face au manque de langage commun, dans un contexte multi-cultural et multidisciplinaire : procédure à
implémenter par l’équipe en AMCD pour éviter des frustrations reliées à ce fait (temps requis pour que tous les
acteurs se comprennent).
233
MODULE AMCD
FACTEURS
Structuration des enjeux (par l’équipe AMCD)
(1) (1)
(2)
TOTAL 1 7 2 1
MODULE AMCD
supplémentaire à faire.
Maitrise de l’outil.
méconnaissances en
tant que non-experts
(risque de ne pas
s’approprier cette
étape).
(1)
235
MODULE AMCD
FACTEURS
Choix de la méthode d’évaluation (par l’équipe d’experts en AMCD)
participation à des
projets internationaux,
capacitation par des
experts chevronnés et
reconnus au niveau
EXTERNE
mondial).
(1) (1)
CONTRAINTES
TOTAL 1 6 1 2
BESOINS/ 1. Méthode d’évaluation choisie en fonction du type de problème analysé (i.e. disponibilité des
DÉFIS acteurs, budget prévu, facilité d’analyse de résultats).
2. La décision appartient au groupe d’experts en AMCD, lequel doit faire preuve d’honnêteté
et d’impartialité.
3. Validation consensuelle de la méthode d’évaluation choisie.
4. Plan scientifique : il faut interpréter plusieurs interfaces.
5. L’expert doit être capable d’en tirer de conclusions.
6. Rendre significatifs les résultats (rédaction d’une grille de lecture sur l’ACP).
7. Mise au point du visuel de D-Sight (information à restituer).
8. Conjoncture démocratique.
9. Appuis financiers (Coûts associés).
10. L’expert doit avoir une position neutre.
236
MODULE AMCD
FACTEURS
Choix de critères (par l’équipe d’experts en AMCD)
Rapidité dans la sélection des critères POUR LES ACTEURS : Étape dans laquelle la
pour passer immédiatement à leur Les critères sont un appui modélisation du problème
pondération. au jugement des enjeux s’établit.
Structuration cohérente des identifiés.
préoccupations des acteurs.
FORCES
Construction du problème.
Il faut démontrer
FAIBLESSES
l’impartialité et l’honnêteté
de l’équipe d’AMCD.
(1)
(1) (1)
EXTERNE
financier. participent.
(1)
(1)
TOTAL 2 4 2 2
BESOINS/ 1. Le choix appartient au groupe d’experts en AMCD, lequel doit faire preuve d’honnêteté et d’impartialité.
DÉFIS 2. Équilibre entre le choix de développer de nouveaux critères, appropriés au contexte spécifique du projet,
et le choix d’utiliser ceux déjà développés.
3. Validation consensuelle des critères choisis.
4. Organisation matérielle (logistique)
5. Identification des liens existants entre les systèmes de valeurs des acteurs et les critères choisis.
6. Appuis financiers (Coûts associés).
237
MODULE AMCD
Authentique représentativité
FAIBLESSES
(1)
238
MODULE AMCD
(1)
CONTRAINTES
TOTAL 1 4 3 2
MODULE AMCD
FACTEURS
Identification des indicateurs (par l’équipe AMCD)
Génération d’information et sa
OPPORTUNITÉS
disponibilité.
(1)
EXTERNE
CONTRAINTES
TOTAL 0 4 1 3
BESOINS/ 1. Le choix appartient au groupe d’experts en AMCD, lequel doit faire preuve d’honnêteté et d’impartialité.
DÉFIS 2. Validation consensuelle, par les acteurs, des indicateurs choisis.
3. Temps alloué à cette validation
4. Information disponible.
5. Conjoncture démocratique.
6. Appuis financiers (Coûts associés, i.e nombre de séances nécessaires).
7. Rendre significatif le choix de l’échelle.
240
MODULE AMCD
FACTEURS
Choix des échelles d’évaluation (par l’équipe AMCD)
Permettra d’évaluer chaque critère POUR LES ACTEURS : Étape dans laquelle les
sur chacun des scénarios. Renforcement des systèmes de experts travaillent de façon
valeurs : graduation d’effets isolée pour créer un
environnementaux, système cohérent qui
FORCES
(3)
CONTRAINTES
TOTAL 0 4 1 1
BESOINS/ 1. Le choix appartient au groupe d’experts en AMCD, lequel doit faire preuve d’honnêteté et d’impartialité.
DÉFIS 2. Validation consensuelle, par les acteurs, des indicateurs choisis.
3. Temps alloué à cette validation.
4. L’échelle n’existe que par les connaissances scientifiques ou du territoire.
5. Information disponible.
6. Appuis financiers (Coûts, i.e nombre de séances nécessaires).
241
MODULE AMCD
FACTEURS
Matrice d’évaluation (équipe AMCD et validée par les acteurs)
Informe plus là-bas de la définition du Base du dialogue des acteurs Synthèse du problème
problème. (dialogue à partir de la même (fait le lien entre la
Représentation achevée du problème matrice d’évaluation). construction du
(indique les performances des Outil de communication avant problème et la
scénarios). d’entrer dans le logiciel (« boîte décision).
FORCES
CONTRAINTES
TOTAL 0 4 3 1
BESOINS/ 1. Le choix appartient au groupe d’experts en AMCD, lequel doit faire preuve d’honnêteté et d’impartialité.
DÉFIS 2. Validation consensuelle de la matrice d’évaluation.
3. Temps alloué à cette validation.
4. Appuis financiers (Coûts associés, i.e. nombre de séances nécessaires).
242
MODULE AMCD
FACTEURS
Analyse de préférences
Obtention des schémas des Immersion dans le dialogue Étape dans laquelle les
critères conflictuels et de la et la négociation. scénarios sont classifiés
coalition des acteurs. Éclaircissement personnel selon leurs performances.
Rangement des scénarios. sur les systèmes de valeurs.
FORCES
Disponibilité d’espaces Avancement de l’état de Débats publics entre les Repérage d’enjeux majeurs
OPPORTUNITÉS
adéquats. connaissances (i.e. santé et opposants et les défenseurs dans la réalisation d’une
ultrasons) (sert à juger la raison d’un projet déterminé. étude d’impact par analyse
d’être des préférences ou de multicritère et de méthodes
préoccupations). de prise de décision.
CONTRAINTES
TOTAL 1 4 3 2
BESOINS/ 1. L’équipe d’experts en AMCD conduit l’analyse et elle doit faire preuve d’honnêteté et d’impartialité.
DÉFIS 2. Conjoncture démocratique.
3. Appuis financiers (Coûts associés).
4. Choix de la fonction de préférence.
5. Défis en communication i.e. présentation de produits sortis du logiciel (effet « boîte noire »).
243
MODULE AMCD
FACTEURS
Consolidation du jugement : Analyse de sensibilité et de robustesse
Indique la plage des paramètres (i.e. le Permets d’éclaircir les Étape dans laquelle se
poids d’un critère) peut changer sans choix personnels sur comprennent les raisons de
affecter le rangement des scénarios, et l’importance accordée à divergence et convergence
vice-versa. chaque critère et, si entre les acteurs.
Aide à identifier quels critères pourront nécessaire, d’y faire des Fait possible une auto-
faire l’objet de la négociation. changements. évaluation du système de
Aide à comprendre les mécanismes de Prise en charge de préférence des acteurs pour
négociation. divergences : comment faire de changements dans
Permets de confronter les divergences on réagirait devant une la pondération de critères.
FORCES
MODULE AMCD
FACTEURS
Consolidation du jugement : Analyse de sensibilité et de robustesse
CONTRAINTES
TOTAL 1 8 3 3
MODULE AMCD
FACTEURS
Décision
Co-construite : avec des acteurs Sentiment de participation Étape dans laquelle il est
autant locaux qu’extérieurs à la dans la prise de décision. possible de structurer la
communauté hôte; avec une Sentiment de refus ou négociation grâce à une
vision partagée du projet de parc d’appropriation de la vision globale et non pas
éolien; en reconnaissant les décision prise. sur un rangement ou sur un
conflits présents; à partir d’une rapport de forces.
nouvelle connaissance générée
par le processus.
Vision partagée des enjeux du
projet.
Mise en évidence des faiblesses
FORCES
cartographie de solution
possible selon les différents
acteurs (plan Gaia et classement
individuel).
MODULE AMCD
FACTEURS
Décision
différents secteurs (i.e. hôpitaux, décision conjointes (i.e. de prise de décision (i.e.
universités, etc.). pratique dans de réseaux dans les études d’impacts).
sociaux sur internet ou à la
participation citoyenne
dans la communauté).
EXTERNE
TOTAL 1 5 3 2
Cette liste présente les facteurs (forces, faiblesses, opportunités, contraintes), par module et par
environnement, apparaissant dans les 16 feuilles de travail (voir annexe F).
26. Possibilité d’obtenir information vectorielle à partir d’un document non disponible en
format numérique.
27. Possibilité de participation multi- acteurs.
28. Possibilité de prise en considération de bases de données de création locale.
29. Prise en compte des critères que chaque acteur défend.
30. Produit « physique » (cartes).
31. Réalisation de la dernière étape du module SIG
32. Réalisation d’une étape du module SIG et préparations d’autres.
33. Réalisation de chacune des étapes du module SIG.
34. Représentation des idées concrètes.
35. Représentation des scénarios à analyser
36. Si existence de produits numériques, accès à information vectorielle facilement
reproductible et qui permet la compréhension du territoire.
37. Validation de l’information et de la qualité de la donnée.
32. Identification de possibles blocages du processus dus à des enjeux non négociables.
33. Immersion dans le dialogue et la négociation.
34. Implication de tous les acteurs.
35. Informe plus là-bas de la définition du problème.
36. Indique la plage des paramètres (c.-à-d. le poids d’un critère) peut changer sans
affecter le rangement des scénarios, et vice-versa.
37. Interrelation des connaissances scientifiques et locales.
38. Les décideurs ont une cartographie de solution possible selon les différents acteurs
(plan Gaia et classement individuel).
39. Maîtrise de l’outil.
40. Mise en évidence des faiblesses et des forces des différents scénarios.
41. Mise en valeur des systèmes de préférences de chaque acteur.
42. Obtention des schémas des critères conflictuels et de la coalition des acteurs.
43. Offre aux acteurs une certitude sur les résultats qui leur sont présentés.
44. Opportunité de réfléchir sur les impacts d’un projet.
45. Outil adapté à la réalité du problème.
46. Outil de communication avant d’entrer dans le logiciel (« boîte noire ») et donc outil
de construction de la transparence.
47. Pas de codification supplémentaire à faire.
48. Permets d’éclaircir les choix personnels sur l’importance accordée à chaque critère
et, si nécessaire, d’y faire des changements.
49. Permets de confronter les divergences entre les acteurs.
50. Phase dans laquelle l’expert travaille de façon isolée pour donner réponse à la
question posée par les acteurs (c.-à-d. la réponse est de classer différentes possibilités d’un même
projet).
51. Permettront d’évaluer chaque critère sur chacun des scénarios.
52. Possibilité de discuter à sujet des indicateurs et des échelles.
53. Possibilité d’utiliser différentes unités (ha, m2, etc.).
54. POUR LES ACTEURS : Les critères sont un appui au jugement des enjeux
identifiés.
55. POUR LES ACTEURS : Lors de la validation, certitude que les indicateurs choisis
représentent les enjeux exprimés (Transparence dans la négociation).
56. POUR LES ACTEURS : Renforcement des systèmes de valeurs : graduation d’effets
environnementaux, socioculturels ou économiques de chaque scénario.
57. Prise en charge de divergences : comment réagirait-on devant une situation?
58. Rangement des scénarios.
59. Rapidité dans la sélection des critères pour passer immédiatement à leur pondération.
60. Rapidité dans la sélection des indicateurs pour passer immédiatement à leur
validation et à la construction de la matrice d’évaluation.
61. Rapidité dans la sélection de la méthode d’évaluation pour passer immédiatement à
l’étape suivante.
62. Reconnaissance de la dimension environnementale, sociale et économique de ses
enjeux.
63. Reconnaissance de l’effet (« positif » ou « négatif ») que chaque enjeu a dans le
système de valeurs des acteurs.
64. Reconnaissance de l’importance que ses systèmes ont dans le dialogue et la
négociation.
65. Reconnaissance objective du contexte auquel font face les acteurs.
66. Remise en question des préférences du problème.
252
Avritzer, L. 2002. Democracy and the Public Space in Latin America. Princeton
University Press. USA.
Blogue énergie éolienne. 2007. Base de connaissances sur l’énergie éolienne avec
blogue, outil de formation en ligne, annuaire, vidéo, document. [En ligne] Page consultée le 4
novembre 2007 : http://blogue-energie-eolienne.org/wp-content/uploads/2007/03/manifeste-
citoyen-pour-un-developpement-eolien-durable1.pdf
Erelia Groupe. 2007. Le haut des ailes. [En ligne] Page consultée le 4 novembre :
http://www.lehautdesailes.fr/
GIRARDOT J.-J. 2009. ‘Results and prospects of caENTI the coordination action of
the Network of Territorial Intelligence’ in CaENTI final seminar in Brussels, Brussels.
Groupe de travail III du GIEC. Mai 2007. Bilan 2007 des changements climatiques :
l’atténuation des changements climatiques. Résumé à l’intention des décideurs. Par Barker
Terry et al. Bangkok, Thaïlande, p. 13.
Midilli, A., Dincer, I. and Murat, A. 2005. Green energy strategies for sustainable
development. Elsevier.
Ressources naturelles Canada. Juillet 2008, page 25. L’acceptation sociale des projets
éoliens au Québec. Par Maya Jegen. Gouvernement du Canada.
Simão, A. et al. 2008. “Web-based GIS for collaborative planning and public
participation : An application to the strategic planning of wind farm sites”. Journal of
environmental Management, doi:10.1016/jjenvman.2007.08.032. Elsevier.
Tsioliaridou, E., Bakos G.C. and Stadler, M. 2005. A new energy planning
methodology for the penetration of renewable energy technologies in electricity factor-
application for the island of Crete. Elsevier.
Annexes
Annexe A
Blackburn, D., et al. 2009. Éoliennes et santé publique. Synthèse des connaissances.
Direction de la santé environnementale et de la toxicologie. Institut national de santé
publique du Québec. https://eolien.ca/fr/bibliotheque/rapports/autres-rapports/item/eoliennes-
et-sante-publique-synthese-des-connaissances.html
Bojorquez-Tapia, L. A. et al. 2002. Integrating Fuzzy Logic, Optimization, and GIS for
Ecological Impact Assessments. Environmental Management Vol. 30, No. 3, pp. 418-433.
Springer.
Canning, G. and Simmons, L. J. 2010. Wind Energy Study – Effect on Real Estate
Values in the Municipality of Chatham-Kent, Ontario. Canadian Wind Energy Association.
http://www.samsungrenewableenergy.ca/sites/default/files/pdf/haldimand/PropertyValuesCo
nsultingReportFebruary42010%5B1%5D.pdf
Cartier énergie éolienne, Pesca Environnement et Hélimax Énergie. 2004. Parc éolien
de Baie-des-Sables. Volume 1. Étude d’impact sur l’environnement. Cartier énergie éolienne.
Colby W. D. et al. 2009. Wind Turbine Sound and Health Effects. An Expert Panel
Review. American Wind Energy Association and Canadian Wind Energy Association.
ttp://www.canwea.ca/pdf/talkwind/Wind_Turbine_Sound_and_Health_Effects.pdf
Côté, F. 2006. Impacts des éoliennes sur les chauves-souris. Direction de la recherche
sur la faune Ministère des Ressources naturelles et de la Faune. Québec.
Failing, L., Gregory, R. et al. 2007. “Integrating science and local knowledge in
environmental risk management: A decision-focused approach”. Ecological Economics
64(1): 47-60. Elsevier.
261
Hoen, B., et al. 2009. The Impact of Wind Power Projects on Residential Property
Values in the United States: A Multi-Site Hedonic Analysis. Office of Energy Efficiency and
262
Renewable Energy. Wind & Hydropower Technologies Program. U.S. Department of Energy
Washington, D.C. http://eetd.lbl.gov/EA/EMP.
Joerin, F. et Musy, A. 2000. ‘Land management with GIS and multicriteria analysis’.
International transactions in operational research 7, 67-78. Pergamon/Elsevier.
Lair, P. 2011, page 5. Vers une stratégie de constitution d'un réseau écologique intra-
forestier à l'échelle du Massif des Bauges. Rapport de stage. Master 2 Professionnel.
Université Aix-Marseille III. Responsable Agathe Leriche. Disponible sur :
http://refora.online.fr/FRENE/PerrineLAIR_2011.pdf (Site Web consulté le 28 mars 2013).
Manzurul Hassan, M., Atkins P. J. and Dunn, C. E. 2003. The Spatial Pattern of Risk
from Arsenic Poisoning: a Banglades Case Study. Journal of Environmental Science and
Health. Vol. A38, No. 1, pp. 1-24. Marcel Dekker, Inc.
263
Maynard, R. et al. 2010. Environmental Noise and Health in the UK. University of
Salford. Health Protection Agency.
Petitclerc, P. et al. 2003. Inventaire pour la protection des espèces végétales menacées
ou vulnérables du milieu forestier. Ministère des Ressources naturelles, de la Faune et des
Parcs. Direction de l’environnement forestier. Direction de la recherche forestière.
http://www.mrn.gouv.qc.ca/publications/forets/connaissances/2003-3091.pdf
Ressources Naturelles Canada. Juillet 2008. L’acceptation sociale des projets éoliens
au Québec. [online] Rapport par Maya Jegen. Gouvernement du Canada. Accessible sur:
http://www.politique.uqam.ca/upload/files/PDF/Rapport_mandate.pdf (Consulté le 25
novembre 2010) (page 25).
Simão, A. et al. 2008. Web-based GIS for collaborative planning and public
participation : An application to the strategic planning of wind farm sites. Journal of
environmental Management, doi:10.1016/jjenvman.2007.08.032. Elsevier.
Sumathi, V.R. et al. 2007. GIS-based approach for optimized siting of municipal solid
waste landfill. Waste Management 28, 2146-2160. Elsevier.
Annexe B
Cartier énergie éolienne, Pesca Environnement et Hélimax Énergie. 2004. Parc éolien
de Baie-des-Sables. Volume 1. Étude d’impact sur l’environnement. Cartier énergie éolienne.
265
Annexe E
Cartier énergie éolienne, Pesca Environnement et Hélimax Énergie. 2004. Parc éolien
de Baie-des-Sables. Volume 1. Étude d’impact sur l’environnement. Cartier énergie éolienne.
Lair, P. 2011, page 5 : Vers une stratégie de constitution d’un réseau écologique
intraforestier à l'échelle du Massif des Bauges. Rapport de stage. Master 2 professionnel.
Université Aix-Marseille III. Responsable Agathe Leriche.