THE ROLE OF MULTIFUNCTIONAL FORESTS IN SUSTAINABLE

LANDSCAPES: A CASE STUDY FROM PORTUGAL

Sónia Maria Carvalho Ribeiro

A thesis submitted for the degree of Doctor of Philosophy

School of Environmental Sciences

University of East Anglia

March 2009

© This copy of the thesis has been supplied on condition that anyone who consults it is understood
to recognise that its copyright rests with the author and that no quotation from the thesis, nor any
information derived therefrom, may be published without the author’s prior, written consent.

1
 THE ROLE OF MULTIFUNCTIONAL FORESTS IN SUSTAINABLE
LANDSCAPES: A CASE STUDY FROM PORTUGAL

Abstract

Forests provide a range of goods and services that are of utmost importance in addressing
the challenge of managing for “sustainable landscapes”. However, it is difficult to
implement sustainable forestry management (SFM) across landscapes because of
interactions between the different scales at which planning and decision making take place,
as well as spatial variations in stakeholder preferences for forest characteristics and the
roles that forest need to perform.

This thesis adopts a multi-scale approach to investigate mechanisms for the

implementation of SFM. It uses GIS techniques together with quantitative (e.g. statistical
analysis) and qualitative approaches (e.g. focus groups meetings), aided by visual tools
such as photographs. The specific focus is on the Portuguese forestry sector with the
following four main objectives: 1) investigate associations between forest characteristics
and socio-economic development at national and regional scales 2) identify the role(s) of
forests within an urban-rural gradient at river basin scale 3) assess variations in public
preferences for forest characteristics at a river basin 4) develop and evaluate scenarios for
SFM in rural parishes within northern Portugal.

The results indicated that forest characteristics, roles and public preferences varied at
several geographical scales. This suggests that strategies for SFM need to vary as well, but
also take into account the linkage of roles across urban-rural gradients. At the parish scale,
two scenario storylines were created, developed and validated by broad group of
stakeholders, but the governance mechanisms to implement these ideas were not in place.

The study provides a template for developing sustainable forestry practices in the
Portuguese context or further afield, but additional research is needed to extend the
“toolkit” used here and address the policy and management challenges that remain.

2
Acknowledgements

This research was made possible by the participation of a great number of people. I am grateful to
all survey respondents, focus group members and workshop participants for their essential
contribution to this study. Special thanks are due to the staff of all the Forestry Municipal offices in
Minho region of Portugal, particularly to those in Ponte da Barca and Arcos de Valdevez and
ARDAL -Associação Regional de Desenvolvimento do Alto Lima -for their assistance in the
scenario creation and development.

I am especially thankful for the invaluable guidance that I have received throughout this project
from my supervisor Professor Andrew Lovett. Without his guidance and encouragement this thesis
could not have been completed.

I would also like to thank other members of the faculty and research staff at the University of East
Anglia who have supported me throughout this journey. Special thanks are due to Professor Tim
O‟Riordan for his guidance and support namely in the implementation of focus groups meetings
and workshop. Thanks are also due to Dr. Peter Simmons, Professor Kate Brown and Professor Ian
Bateman. Special thanks are also due to Trudie Dockerty, Katy Appleton, Janice Darch, Elah Matt
and Adrian Southern for their help in proof reading the chapters of this thesis. I am particularly
grateful to my field assistants at different stages of the project namely David Benson, Thunuadee,
Tania Teixeira, Linda Cerqueira, Cristina Vieira, Beatriz Contreras and Gil Ferreira.

Huge thanks are also due to my friends and colleagues with whom I have shared much of my time
at the University of East Anglia, namely Marta, Irena, Natália, Beatriz, Janice, Ana, Joana, Zé,
Angelo, Marco, Nunos, Elah, Adrian, Camy, Saffron, Pham, Steve. The Portuguese-international
lunches in which we all participated every Thursdays were undoubtedly a great moment of the
week.

Funding for this research was provided by the Fundação para a Ciencia and Tecnologia in Portugal
to whom I am sincerely grateful.

I wish to thank to my family and friends for being always with me despite from afar. Special
thanks go to my mother Gemina, my father João my sisters and brother Augusta, Fátima and
Serafim and their respective partners Teixeira, Augusto and Sameiro as well as to my nieces and
nephews Tania, Tina, Sofia, Miguel, João and David. To all my friends who shared with me the
great experience of being alive and certainly shaped who I am today.

Finally, I thank my husband, Gil, for his support throughout this journey. Above all, thank you for
bringing happiness to my live.

3
Abbreviations

C& I Criteria and Indicators

CAP Common Agriculture Policy

EA Ecosystems Approach

EU European Union

NGO Non Governmental Organisation

SD Sustainable Development

SES Social- Ecological System

SFM Sustainable Forestry Management

WFD Water Framework Directive

4
 TABLE OF CONTENTS

CHAPTER 1. INTRODUCTION: SUSTAINABLE LANDSCAPES & MULTIFUNCTIONAL FORESTS ......................... 8

1.1. FORESTS IN LANDSCAPES ............................................................................................................................... 9
1.1.1. Sustainable Development (SD)........................................................................................................ 9
1.1.2. Pursuing SD at the landscape scale .............................................................................................. 11
1.1.3. Forests, landscapes and sustainable development ....................................................................... 13
1.1.4. Defining the problem .................................................................................................................... 17
1.2. SUSTAINABLE FORESTRY MANAGEMENT & MULTIFUNCTIONAL FORESTS ................................................................ 19
1.2.1. Sustainable Forestry Management (SFM) .................................................................................... 20
1.2.2.Multifunctionality .......................................................................................................................... 25
1.2.2.1. Multifunctionality: the forest scale ........................................................................................................ 26
1.2.2.2. Goods and services provided by forests ................................................................................................ 27
1.2.2.3. Multifunctional forestry vs. dominant use ............................................................................................. 34
1.2.2.4. Good forestry practices ......................................................................................................................... 35
1.2.3. Multifunctional landscapes ........................................................................................................... 38
1.3. SUSTAINABLE LANDSCAPE PLANNING AND LANDSCAPE GOVERNANCE AS A WAY TO MOVE FORWARD........................... 41
1.4. RESEARCH RATIONALE AND AIMS .................................................................................................................. 49
1.5. THESIS STRUCTURE ..................................................................................................................................... 51

CHAPTER 2. FORESTRY AND LAND MANAGEMENT IN EUROPE: A CASE STUDY FROM PORTUGAL ............ 53
2.1. FORESTRY IN EUROPE ................................................................................................................................. 54
2.2. A CASE STUDY IN PORTUGAL ........................................................................................................................ 60
2.2.1. Brief description of the country .................................................................................................... 60
2.2.2. Forestry and land use planning in Portugal .................................................................................. 63
2.2.3. Moving towards SFM in Portugal ................................................................................................. 69
2.2.4. Research questions ....................................................................................................................... 72

CHAPTER 3. ASSOCIATIONS BETWEEN FOREST CHARACTERISTICS AND SOCIO-ECONOMIC DEVELOPMENT:

A CASE STUDY FROM PORTUGAL .............................................................................................................. 75
3.1. INTRODUCTION ......................................................................................................................................... 77
3.2. DATA AND METHODS ................................................................................................................................. 80
3.2.1 Data sources .................................................................................................................................. 80
3.2.2. Analysis techniques ....................................................................................................................... 81
3.2.2.1. Ranking method ..................................................................................................................................... 82
3.2.2.2. Principal Component Analysis (PCA) and Cluster Analysis (CA) ............................................................. 83
3.3. RESULTS .................................................................................................................................................. 84
3.3.1. National scale trends .................................................................................................................... 84
3.3.2. Regional scale trends .................................................................................................................... 85
3.3.2.1. Socio-economic classification ................................................................................................................ 87
3.3.2.2 Variations in class metrics ....................................................................................................................... 89
3.4. DISCUSSION .............................................................................................................................................. 92
3.5. CONCLUSIONS ........................................................................................................................................... 95

5
CHAPTER 4. INTEGRATING PUBLIC USES AND PREFERENCES IN THE DESIGN OF MULTIFUNCTIONAL PLANS
AT THE CATCHMENT SCALE: A CASE STUDY IN THE MINHO REGION OF PORTUGAL .................................. 96
4.1. INTRODUCTION ......................................................................................................................................... 98
4.2. DATA COLLECTION AND METHODS ..............................................................................................................105
4.2.1. Study area ...................................................................................................................................105
4.2.2. Overall method and approach ....................................................................................................107
4.2.3. Characterisation of forest plots in the study area ......................................................................109
4.2.4. Questionnaire survey ..................................................................................................................110
4.3. RESULTS ................................................................................................................................................111
4.3.1. Characterisation of forest plots in the study area ......................................................................111
4. 3.2. Results of the questionnaire survey ...........................................................................................113
4.3.2.1. Uses of forests ..................................................................................................................................... 113
4.3.2.2. Preferences for management strategy ................................................................................................ 116
4.3.2.3 Trends of change in forest characteristics ............................................................................................ 120
4.3.3. Comparing the field and questionnaire surveys..........................................................................121
4. 4. DISCUSSION ...........................................................................................................................................126
4.5. CONCLUSION ..........................................................................................................................................129

CHAPTER 5. PUBLIC OPINION REGARDING ATTRACTIVENESS AND MANAGEMENT OF FOREST

LANDSCAPES: PREFERENCES FOR FOREST COVER AND STAND STRUCTURE ............................................ 131
5.1. INTRODUCTION .......................................................................................................................................133
5.2. MATERIAL AND METHODS .........................................................................................................................137
5.2.1. Study area ...................................................................................................................................137
5.2.2. Questionnaire survey ..................................................................................................................137
5.2.2.1. The questionnaire images .................................................................................................................... 138
5.2.2.2. Statistical analysis ................................................................................................................................ 139
5.3. RESULTS ................................................................................................................................................140
5.3.1. OVERALL RESULTS .................................................................................................................................140
5.3.2. Correlation coefficients of attractiveness and management rankings .......................................143
5.3.2.1. Percentage of forest cover (F photos) ................................................................................................. 143
5.3.2.2. Stand structure (S photos) ................................................................................................................... 145
5.3.3. Attractiveness vs. management across user groups ...................................................................146
5.3.3.1. Type of users ........................................................................................................................................ 147
5.3.2.2. Attractiveness vs. management preferences across the rural/urban gradient ................................... 148
5.3.3. Public preferences for forests and ecology of forests ecosystems ..............................................151
5.3.4. Contrasting verbal and visual approaches ..................................................................................151
5.4. DISCUSSION ............................................................................................................................................153
5.5. CONCLUSIONS .........................................................................................................................................156

6
CHAPTER 6. GOVERNING FOR SUSTAINABILITY: IMPLEMENTING SUSTAINABLE FORESTRY MANAGEMENT
IN NORTHERN PORTUGAL ....................................................................................................................... 157
6.1. INTRODUCTION .......................................................................................................................................159
6. 2. STUDY AREA AND METHODS ......................................................................................................................164
6.2.1 Study area and case study selection ............................................................................................164
6.2.2. Forestry and land management in the study area ......................................................................165
6.2.3. Creation and development of scenario storylines .......................................................................167
6.2.3.1. Focus group meetings .......................................................................................................................... 169
6.2.3.2. Land use change models ...................................................................................................................... 171
6.2.4. Scenario evaluation and implementation ...................................................................................173
6.2.4.1. Workshop............................................................................................................................................. 173
6.2.4.2. Implementation ................................................................................................................................... 175
6.3. RESULTS ................................................................................................................................................175
6.3.1. Description of the storylines .......................................................................................................178
6.3.2. Scenario evaluation and implementation ...................................................................................181
6.4. DISCUSSION ............................................................................................................................................185
6.5. CONCLUSION ..........................................................................................................................................187

CHAPTER 7. CONCLUSIONS: THE ROLE OF MULTIFUNCTIONAL FORESTS IN SUSTAINABLE LANDSCAPES . 188

7.1. HOW CAN SUSTAINABLE LANDSCAPE PLANNING AND LANDSCAPE GOVERNANCE GUIDE FOREST PLANNING AND
MANAGEMENT? ............................................................................................................................................. 189
7.2. WHAT ROLES MIGHT MULTIFUNCTIONAL FORESTS HAVE IN SUSTAINABLE LANDSCAPES? .........................................195
7.3. THE ROLE OF MULTIFUNCTIONAL FORESTS IN SUSTAINABLE LANDSCAPES: WIDER IMPLICATIONS ...............................201
7.3.1. European forestry will become more diverse in the future .........................................................204
7.3.2. The need for a portfolio of measures for SFM in Europe ............................................................205
7.3.3. Enhanced communication between researchers, the public and decision makers is important 206
7.4 RESEARCH CAVEATS AND FURTHER RESEARCH .................................................................................................207
7.4.1. Research caveats: problems and difficulties ...............................................................................207
7.4. 2. Future research: there is no panacea but there might be trends...............................................210
7.5 CONCLUDING REMARKS..............................................................................................................................211

REFERENCES ............................................................................................................................................ 212

APPENDICES ............................................................................................................................................ 234

7
 ...We came to see that a new development path was
required, one that sustained human progress not just
in a few places for a few years, but for the entire
planet into the distant future” WCED (1987:4)

CHAPTER 1. INTRODUCTION: SUSTAINABLE LANDSCAPES &

MULTIFUNCTIONAL FORESTS

8
1.1. Forests in landscapes
The research reported in this thesis concerns the issue of how to plan and govern forests in
a sustainable manner. Sustainable development (SD) needs addressing at different
ecological and institutional scales (Cash et al., 2006; CBD, 2008; O' Riordan and Voisey,
1998; OECD, 2002, 2006). Recent work has stressed the need for pursuing SD at the
landscape scale because it is at this scale that stakeholder demands influence landscape
development (Naveh, 2007; Selman, 2006; Tress et al., 2001). Despite recognising
landscapes as an appropriate scale to deal with sustainability there are issues that need to
be carefully addressed in order to attain sustainable forestry management within a
sustainable landscape framework. These issues set the frame for the general problem
addressed throughout this thesis.

Section 1.1 of this chapter defines the problem under analysis, whilst Section 1.2 explores
the background literature regarding both Sustainable Forestry Management (SFM) and
multifunctional forestry. The third section explores the ways in which sustainable
landscape planning and landscape governance can guide the development of strategies to
implement sustainable forestry management across landscapes. Finally, Section 1.4 sets
out the main aims of the research, and the multi-scale approaches that were developed to
inform planning and management strategies for SFM at different spatial scales.

1.1.1. Sustainable Development (SD)

There has been an extensive discussion in the literature about the integration of socio-
economic and environmental issues in order to implement sustainable development (SD) at
spatial scales from the international to the local (Kozlowski and Hill, 1993; OECD, 2002,
2006; Sachs, 2004). SD has challenged governments and societies as well as science to
work together (Cash et al., 2003), embracing the challenge of putting into place
development strategies able to meet the “needs of the present without compromising the
ability of future generations to meet their own needs” (WCED, 1987:8). Although
contested and several times redefined (e.g. UNEP) the term has been central to the
international environmental debate. Pursuing SD implies integrating societies, economies
and environment in decision making and participation in development processes
(Cashman, 2006; Jabareen, 2008). It implies not some form of present status quo but rather
the potential transformation and evolution of the current economic and social paradigm
towards the integration of environmental concerns in development processes (Cash et al.,
2003; O' Riordan and Stoll-Kleemann, 2002)

9
In order to deal with the inherent complexity of pursuing SD a set of concepts have been
developed that help to set the frame for moving towards it (McMichael et al., 2003).
Managing natural resources without undermining the natural capital of an area (Costanza
and Daly, 1992) is seen as a prudent “rule” for assuring sustainability. Natural capital is
often interpreted as the stock of natural ecosystems that yields a flow of valuable
ecosystem goods or services into the future (Balmford et al., 2002; Costanza et al., 1997).
Since the concept appeared policies are often classified as “weak” or “strong”
sustainability dependent on the extent to which they assure the natural capital rule
(Costanza and Daly, 1992).

Pursuing SD also involves enhancing social capital which refers to connections within and
between social networks as well as connections among individuals (Pretty, 2003; Pretty
and Ward, 2001). Another concept which has helped to tackle the SD challenge is that of
social-ecological systems (SESs) (Folke et al., 2002; Folke et al., 2005). These social-
ecological systems can be seen as a set of interactions between the human system
(characterised by both structure and functioning e.g. institutions) and ecosystems (that also
have their own structure and functioning). For human populations embracing SD means
transforming our way of living to increase the chances that environmental conditions will
indefinitely support human well-being by continually assuring the flow of non
substitutable goods and services from ecosystems (McMichael et al., 2003). As a
consequence, ecosystems are increasingly seen as capital assets, with the potential to
generate a stream of vital life-support services meriting careful evaluation and investment
(Turner and Daily, 2008).

Ecosystems can be defined at a wide range of scales ranging from the level of a plant up to
biomes covering several thousand square kilometres (Box 1.1). In general, large scale, long
period phenomena set physical constraints on smaller scale, shorter period ones (MEA,
2003). In the same way, large scale processes may be driven by the joint impact of small
scale processes (Folke et al., 2005). For example, microbes operate on the scale of
micrometers and minutes but their cumulative activity determines large scale processes
such as the nutrient cycle (e.g. demineralisation of organic material and nitrogen fixation).

In parallel with ecological scale, in social-ecologycal systems (SESs), a hierarchy of

institutions can be distinguished (Ostrom, 2007). The different scales reflect the different
levels at which decisions on the utilization of resources are made. At the lowest
institutional level, this may include individuals or households (Box 1.1). Higher
institutional scales such as the provincial, national or international can be also
distinguished (O' Riordan and Voisey, 1998). Similarly to ecosystems, there are also cross
10
scale interactions between institutional levels in such a way that decisions made at one
scale influence the scales above and below (CBD, 2008). For example the EU, through the
Common Agriculture Policy (CAP) (International scale), influences the ways in which land
is managed for agriculture at the household scale (Deybe, 2007).

Box 1.1 Ecological and institutional scales

Ecological scales Institutional scales

Global International

Biome National

Landscape State/provincial

Ecosystem Municipal

Plot Household

Plant Individual

Source: Adapted from de Groot and Hein (2007: 30)

Moving towards SD is an overarching goal worldwide (Cash et al., 2003; O' Riordan and
Stoll-Kleemann, 2002; OECD, 2002). However, when the balance between economy-
society and environment has been attempted in practice the concept has been revealed to
be sufficiently “open-ended” to leave a vast array of problems to be solved (Jacobs, 1986).
Approaches to tackle SD have to a large extent focussed on i) single issues or resources not
properly addressing it ii) scale dependency (Folke et al., 2005). Traditionally, disciplines
have addressed SD in a “specialised” manner by dealing separately with fields such as
(sustainable) agriculture, (sustainable) forestry (Siry et al., 2005), (sustainable) cities,
(sustainable) water management (Cashman, 2006), without focussing on the interactions of
these systems in the wider issues related to the landscape (Lindenmayer et al., 2008).

1.1.2. Pursuing SD at the landscape scale

Landscape was defined in the European Landscape Convention as “an area, as perceived
by people, whose character is the result of the action and interaction of natural and/or
human factors” (Council of Europe, 2000). Landscape is a concept of multiple meanings. It
results from the ways that different components of the environment both ecological
(geology, soils, climate, flora and fauna) and cultural (institutions) interact together in
material and imaginary ways (Selman 2006:13). Modern theories of landscape represent it
as a holistic1 entity within which natural and human processes merge, and where economic

1
Holism means that the whole is more important than the composing parts. Holistic also means that each
element receives its significance only because of its position and relationship with the surrounding elements
in Antrop and Eetvelde (2000)
11
social and ecological objectives can be balanced (Antrop, 2005, 2006; Jacobs, 1986;
Selman, 2006). Furthermore, in a sustainable landscape a multitude of functions occur
simultaneously and link people and environment altogether in a self-reinforcing manner
(Antrop, 2005, 2006; Jacobs, 1986; Mander et al., 2007; Selman, 2006). According to
Forman (1995:519) a sustainable environment “is an area in which ecological integrity
and basic human needs are concurrently maintained over generations”. Based on the two
previous definitions (landscape and sustainable environment) it can be said that developing
strategies2 for sustainable landscapes involves guiding the process of interaction between
natural (e.g. ecosystems) and human factors (e.g. institutions) in ways that landscapes
(seen as the area as perceived by people) will neither lose ecological integrity nor the
ability to fulfil basic human needs.

The broadening of the SD concept to a landscape level combines a focus on sustainability

of ecosystems (e.g. forests) and places (e.g. cities, rural areas) in such a way that
sustainability is pursued through landscapes (Lindenmayer et al., 2008; Selman, 2006). It
does not mean that pursuing sustainability of a single ecosystem (e.g. fisheries or forestry)
or area (e.g. rural) is less valued, instead, it means that sustainability may be better
addressed, and more realistically implemented, through a landscape perspective (Potschin
and Haines-Young, 2006b; Selman, 2006). This is more realistic in the sense that
humankind does not interact with either forests or agriculture in isolation, neither are rural
areas or cities separate entities, instead there are a multitude of interactions between them.
Therefore, there is a need to put into practice integrated natural resource management
(INRM) which incorporates multiple aspects of use of different natural resources (e.g.
forests, fisheries) into a system of sustainable management to meet explicit production
goals as well as other goals of the wider community (Sayer and Maginnis, 2005).

In addition, there are at present strong arguments to examine issues of environmental

condition (e.g. biodiversity) at spatial scales3 larger than a single ecosystem recognising
that the “dynamic whole is greater than the sum of the parts” (Antrop, 2006; Lindenmayer
et al., 2008; Lindenmayer, 1999; Liu and Taylor, 2002). It is increasingly acknowledged
that depending on the spatial scale under analysis the number of ecosystem functions that
can be observed varies: in larger areas the full spectrum of ecosystem functions can be
realised (e.g. carbon sequestration, biodiversity), whereas at the level of one field fewer

2
Strategies are proactive, based on a plan, intended to effect the forces (causes) of conflicts and problems
Ahern (1995)
3
Spatial scale refers to the spatial dimension of an object or process-that is by the degree of resolution and
the size of the geographical area in question. Adapted from Selman (2006:24)
12
functions (e.g. productivity) appear (Lindenmayer et al., 2008; Mander et al., 2007; Moore
et al., 2009).

1.1.3. Forests, landscapes and sustainable development

Forestry is one activity that needs to be addressed and coordinated at a multitude of scales
(CBD, 2008; MCPFE, 2007) because the ways in which goods and services are provided
and delivered in forestry also occur across a range of spatial scales (e.g. carbon
sequestration, biodiversity) (Box 1.1). Consequently, sustainable forestry management
SFM (the concept is defined in Section 1.2) needs to be coordinated both at different
ecological and institutional scales (Box 1.1) (Baskent and Yolasigmaz, 2000; Olivier et al.,
2000).

Forests are crucial to conserve biological diversity, water resources, soils and to maintain
landscape ecological functions and ecosystem integrity (FAO, 2003; MCPFE, 2007; MEA,
2003). Due to the ability of forests to provide an array of functions i) at different scales and
ii) in different places within the landscape (Section 1.2), the role of forestry has been
increasingly recognised to be of utmost importance in sustaining landscapes (FAO, 2003;
MCPFE, 2007; MEA, 2003; Spiecker, 2003; Stengera et al., 2009). Compared with other
land use types such as agriculture, trees and forests have a longer permanency in
landscapes and the benefits from forests extend far away from the places where forests are
located (e.g. upland forests protect lower areas from soil erosion) (Fisher et al., 2004;
Sayer and Maginnis, 2005). Section 1.2 further explores this topic.

Communities, legislators, industry, local stakeholders and the public at large make
different demands on landscape and influence landscape development (McMichael et al.,
2003). Consequently, yet recognising the cross-scale effects, it is at the landscape level
that conflicting interests of production and conservation need to be coordinated (Tress et
al., 2001), thus it is at the landscape scale that SD may be pursued and its implementation
tested (Selman, 2006). This work focuses mainly on cultural and protected landscapes.
Definitions of these two landscape types are given in Box 1.2.

Box 1.2. Landscape types

Cultural landscape “...areas whose extent people intuitively grasp and whose distinctive
character derives from centuries of human activity” (Selman 2006:7)
Protected (areas) landscape “...areas of land,.., where the interaction of people and nature
over time has produced an area of distinct character with significant aesthetic, ecological,
and/or cultural value, and often with high ecological biodiversity (IUCN, 1994 cited by
Selman 2006:6)

13
Nature conservation practice has long been defensive, focussing on the protection of nature
reserves on the one hand, and the preservation of particular species on the other (Duhme et
al., 1997). However, as research is demonstrating, nature conservation will be better
addressed if wider conservations strategies across both cultural and protected areas are put
into place. Therefore, conservation goals require strategies for managing the “whole
landscape” (concept defined in Section 1.3.) (Duhme et al., 1997; Margules and Pressey,
2000).

Within the “whole landscape” there are rural and urban areas and sustainability should deal
with both urban and rural issues (Antrop, 2006). If it is likely that sustainable urbanized
landscapes will imply completely different aspects than sustainable rural ones, it is also
likely, that rural and urban “sustainabilities” will depend on each other (Antrop, 2006). The
theme of “urban-rural” partnerships, as addressed by the European Spatial Development
Perspective (ESDP), incentivises cooperation between towns and countryside aiming at
strengthening functional regions (CEMAT, 2007; Selman, 2006).

There are different types of landscapes and one of the factors varying between them is the
amount and type of forest ecosystems. However, trees, woodlands and forests are a
recurrent feature in many landscape types, independently of whether they are protected,
cultural, urban or of rural character (Sayer and Maginnis, 2005). So, it is at the landscape
level that different interests or multiple preferences for forest management need to be
tackled (Sayer and Maginnis, 2005). It is important however to acknowledge that because
forests are one landscape feature they may be seen as only one aspect in addressing
landscape sustainability.

Acknowledging the potential of the landscape approach has lead to a shift from focussing
on sustaining individual ecosystems (e.g. agriculture and forestry) and places (e.g. cities,
rural areas) to embracing the challenge of sustainable landscapes (Forman, 1995; Selman,
2006). Increased awareness of the importance of managing natural resources i) in an
integrated manner and ii) at broader spatial scales (Moore et al., 2009) has lead to the
development of a set of strategies by the European Union (EU) such as the “Pan-European
Biological and Landscape Diversity Strategy”, Water Framework Directive, European
Landscape Convention and Carbon markets (Box1.3).

14
Box 1.3. Policies that encourage a landscape approach

Water Framework Directive In 2000 EU launched the Water Framework Directive (WFD)
which sets up the future frame for regulation and protection of water resources in Europe.

European Landscape Convention Promotes the protection, management and planning of

European landscapes and organises European co-operation on landscape issues. It is the first
international treaty to be exclusively concerned with all dimensions of European landscape

Pan-European Biological and Landscape Diversity Strategy (PEBLDS) The principal aim of
the Strategy is to find a consistent response to the decline of biological and landscape diversity in
Europe and to ensure the sustainability of the natural environment

Carbon market The global carbon market established in 2001 as part of the Kyoto protocol
allow the trade of carbon credits. In January 2005 the European Union established the
Greenhouse Gas Emission Trading Scheme (EU ETS)

Source: Council of Europe webpage. http://ec.europa.eu/environment [Accessed 12

February 2009]

Across a variety of landscapes different socio ecological systems (SESs) are likely to
“explore” different functions of forest ecosystems (the concept of forest functions is
defined in Section 1.2). Some clearly put emphasis on the productive functions, for
example timber production. Other communities explore non timber products such as
livestock grazing or even recreational use (Janse and Ottitsch, 2005). Increasingly, others
are willing to get into carbon markets (Basu, 2009) or focus on transforming forest
biomass into “green sources of energy” (Okkonen, 2008). Recently, forest managers,
governments and societies have become engaged in a constant trade off between
productive and other environmental functions such as protection and recreation (Sayer and
Maginnis, 2005). The problem is that management strategies undertaken in different places
across landscapes influence the delivery of services from forests i) at a variety of scales
and ii) in places other than where management strategies occur (Section 1.2).

It also has been reported that across landscapes public preferences for forests vary hugely.
A review of public preferences for forests indicates that preferences are likely to vary by
personal characteristics and socio-economic factors such as gender and landownership
(Abello and Bernaldez, 1986; Tips and Vasdisara, 1986), professional background (Rogge
et al., 2007b; Winter, 2005) or type of recreational activity (Harshaw et al., 2006; Roovers
et al., 2002). There are also studies reporting differences and conflicts between
preferences of “local and extra-local” inhabitants of protected areas (Zube, 1986).
Therefore, there is a need to incorporate a multitude of public preferences into forest
management (Sheppard and Harshaw, 2000, 2001).

Another issue is that a huge amount of research done so far, for example by Elianor
Ostrom and colleagues, shows that there is no panacea in addressing natural resource
15
management in different SESs (Ostrom, 2007; Ostrom et al., 2007). The multiple processes
occurring in complex SESs (of which forests are only a part) raise enormous challenges in
the implementation of SD and hence of sustainable forestry management (SFM) in
different landscape types (Box 1.2). It is known that strategies to implement SD are likely
to vary according to socio-economic and environmental circumstances making it likely that
there is more than one trajectory to, or scenario for, a sustainable landscape (Antrop, 2006;
Sayer and Maginnis, 2005). Moreover, it has been reported that even in places with
identical socio-economic characteristics and similar environmental conditions the
development paths can greatly differ (Niskanen and Lin, 2001). So, there is a need to go
beyond panaceas and explore different solutions to specific contexts for integrating forests
with other land use types within the landscape.

Different ecosystems combine in a variety of mosaics of land use4/land cover5 mosaics to

create heterogeneous landscapes (Fry, 2001). Consequently, heterogeneity is a basic
characteristic of landscape implying the capacity of landscape to support various,
sometimes contradictory functions simultaneously (Antrop, 2004, 2006). Many elements in
either protected or cultural landscapes (Box 1.3) have a multifunctional character (Mander
et al., 2007; Tress and Tress, 2001, 2003). For example, hedgerows and forests as well as
various agricultural and grassland ecosystems control various energy and material fluxes in
the landscape which simultaneously protect biodiversity and provide both income and
recreational opportunities for people (Matsuoka and Kaplan, 2008; Matthews and Selman,
2006).

During the late 20th century landscape functions such as the production of agro-forestry
products or nature conservation, have tended to become segregated in most European
landscapes as a result of specialisation and intensification of production (Antrop, 2004,
2005). Selman (2006:15) considers “this functional separation of land to be an underlying
contributor to many environmental problems”. In these cases, new associations (or
reinventing old ones) between people, places and economies are to be welcomed if they
create functionality and coherence.

In the same way that landscape multifunctionality is at peril in some areas so the
multifunctionality of agriculture and forestry is not self reinforcing (Pereira et al., 2005;
Pereira and Fonseca, 2003). In countries such as Spain and Portugal for example, the agro-

4
Land use corresponds to the socio-economic description (functional dimension) of areas: areas used for
residential, industrial or commercial purposes, for farming or forestry, for recreational or conservation
purposes. Above all is the purpose of human activity on the land. In European Environment Agency
http://glossary.eea.europa.eu/EEAGlossary/L/land_use
5
Land cover refers to the vegetation, structures, or other features that cover the land e.g. grass, forests,
16
silvo systems are being threatened by several socio-economic changes (Andresen and
Castelbranco, 1993; Firmino, 1999). Consequently, there is a need to study the ways in
which agriculture and forestry may be able to contribute to sustainable landscapes. This
calls for the study of “viable” multifunctionalites (Pinto-Correia and Breman, 2008)
according to different vocations of the territories as well as the social dynamics in place.

In the forestry sector there is a need for integration of different functions in order to fully
address the goal of managing forests sustainably (Section 1.2 and Chapter 4). There is also
a need to find ways in which multifunctional forestry may contribute to more sustainable
landscapes (Sheppard and Harshaw, 2001; Siry et al., 2005). This implies going beyond
describing the “idyllic” roles of forests from the past. As was previously stated, the role of
forestry is changing from not only productive, but also consumptive (e.g. recreation
activities in aesthetically pleasing forest areas) and protective (e.g. providing biodiversity,
flood avoidance and soil erosion) and this will require a new coordination of forest
functions in post-industrial landscapes that needs to be carefully addressed (Nabuurs et al.,
2001; Niskanen and Lin, 2001; Slee, 2007a). This “new” holistic approach does not
physically and permanently divide the land base for certain uses, instead, it aims to manage
the forest ecosystems as a whole to provide multiple functions simultaneously (Baskent
and Yolasigmaz, 2000). Managing forests through a landscape perspective implies
addressing multiple functions from forests (multifunctionality is further explored in
Section 1.2) as it has to include multiple demands from different stakeholders also
reflecting their multiple preferences regarding forests.

1.1.4. Defining the problem

It follows from the above that despite agreement about the (potential) role of forestry in
pursuing sustainable development there are difficulties in the implementation of
sustainable forestry management (SFM) across a range of landscape types because:

1. The landscape scale is influenced by and, in turn, influences other scales (Box 1.1),
2. At the landscape scale there are a multitude of users with different interests in
forest management
3. There are also different preferences for forest characteristics amongst stakeholder
groups within the landscape,
4. The amount and type of forests varies across landscapes and thus the role of forests
in sustainable landscapes is also likely to differ.

Consequently, across landscapes, the ways in which strategies can be developed for
managing the whole mosaic of ecosystems (of which forests are only a part) in ways that
17
landscapes as “the area as perceived by people” will neither lose their ecological integrity
nor the ability to fulfil basic human needs has been receiving much attention (Ahern, 2005;
Antrop, 2005, 2006; Blaschke, 2006; Lindenmayer et al., 2008; Tress et al., 2001) .

Though recognising landscapes as an appropriate scale to deal with sustainability issues the
ways in which SD may be achievable through landscapes is an ongoing debate . Blaschke
(2006:198) refers to the developments concerning “sustaining landscapes” as follows:

“Environmental management has predominantly focussed on individual ecosystems but is

increasingly confronted with managing and planning entire landscapes which often consist
of complex interacting mosaics of different habitat patches and ecosystems...those interact
and combine in a way that may not be obvious...”

As sustainable development is based on the transformation of living resources to achieve

societal goals, planning and governance mechanisms must deal with managing that process
of transformation (or paths) in the interests of long term sustainability (Jacobs, 1986;
Naveh, 2007). Independently of the process of transformation/path to be followed for a
transition to SD to occur, planning and management enforced by effective governance
systems (concepts defined in Section 1.3) are recognised to be key vehicles to deliver SD
through a range of world landscapes (Ahern, 2000, 2005; Hanna, 2005; Selman, 2006).
Moreover, planning and governance systems address and coordinate mechanisms (e.g.
plans, policies) at a multitude of scales (Box 1.1) in such a manner that there is cross-scale
coordination to allow the implementation of SD through landscapes (Blaschke, 2006;
Naveh, 2007; von Haaren and Ott, 2008).

Antrop (2006:195) reinforces the need for planning to achieve sustainable landscapes
“Sustainable landscapes are no fiction if the landscapes qualities are well defined and the
context of change and future functioning is set right and fixed.”

Studies such as that by Forman and Collinge (1997) further demonstrate that nature is best
conserved with rather than without planning as a tool to guide the process of change.

Summarising, this section has described the advantages of pursuing SD through a

landscape perspective and the difficulties in reconciling a multitude of aspects of forest
management through landscapes. Following this, the role of planning and governance
systems has been stressed as of utmost importance in order to guide the interactions
between humans and nature. Concerning forest management, guidance from planning and
governance systems, needs to be coordinated at a range of scales due to the nature of
services from forests (Section 1.2). One of the issues is how to reconcile a multitude of

18
forest functions which reflect a multitude of interests into management strategies where
there are different preferences for forest characteristics.

The problem, then is, how to guide sustainable forestry management at a range of scales in
order to move towards sustainable landscapes. Moreover, what might be a set of tools and
methods (based on planning and governance systems) that could help to guide the
implementation of sustainable forestry management at different spatial scales (see Section
1.3). In addition, there is a need to address the multiple roles that forests may have in
sustainable landscapes (Chapter 2). Box 1.4 summarises these problems which are
addressed throughout this thesis.

Box 1.4. Framing the general problem/general questions

A. There is a need to address forest management at different spatial scales- multiscale

approach.
B. There are a multitude of users of forest resources that need to be reconciled at the
landscape scale
C. Across landscapes there are differences in public preferences for forests and these
need to be addressed in forestry planning
D. Planning and governance systems are seen as the way to move forward in order to
“guide the process” of interaction between man and nature
General problem/questions:
1. How to guide the implementation of SFM across a range of scales? What are
the set of tools and methods (from sustainable landscape planning and
landscape governance) that can help to guide the implementation of SFM at
different spatial scales?
2. What role(s) forests might have in a sustainable landscape? What type of cross-
scale coordination will be likely to deliver the implementation of SFM at the
landscape scale in the context of Portugal?
3. What are the wider implications of this work?

Section 1.3 explores the ways in which planning and governance might inform strategies
able to deliver more sustainable landscapes. It also reviews the ways in which concepts and
tools from landscape planning/governance approaches might help to guide forestry
management with sustainable goals at different scales. However, before addressing those
issues Section 1.2 defines the concepts of sustainable forestry management and
multifunctionality.

1.2. Sustainable forestry management & multifunctional forests

This section begins by defining the concept of sustainable forestry management, aiming at
describing the ways in which forests may contribute to more sustainable landscapes.
19
Following this, the concept of multifunctionality is explored at the forest scale (Section
1.2.2) as well as at the landscape scale (Section 1.2.3). At the forest scale emphasis is put
on the description of forest functions and the ways they may be delivered through
landscapes.

1.2.1. Sustainable Forestry Management (SFM)

Different definitions of the term “forest” are used in different contexts. The definition used
by the Food and Agriculture Organisation (FAO) after consultation with experts worldwide
in 2002 is as follows:

“Land with tree crown cover (or equivalent stocking level) of more than 10 percent and
area of more than 0.5 hectares (ha). The trees should be able to reach a minimum height of
5 meters (m) at maturity in situ. May consist either of closed forest formations where trees
of various storeys and undergrowth cover a high proportion of the ground; or open forest
formations with a continuous vegetation cover in which tree crown cover exceeds 10
percent. Young natural stands and all plantations established for forestry purposes which
have yet to reach a crown density of 10 percent or tree height of 5 m are included under
forest, as are areas normally forming part of the forest area which are unstocked as a
result of human intervention or natural causes but which are expected to revert to forest. It
also includes: forest nurseries and seed orchards that constitute an integral part of the
forest; forest roads, cleared tracts, firebreaks and other small open areas; forest in
national parks, nature reserves and other protected areas such as those of specific
scientific, historical, cultural or spiritual interest; windbreaks and shelterbelts of trees
with an area of more than 0.5 ha and width of more than 20 m; plantations primarily used
for forestry purposes, including rubber wood plantations and cork oak stands.”

(Definition by FAO 2002 available at

http://www.fao.org/docrep/006/ad665e/ad665e06.htm ) [Accessed online 10 January 2009]

Throughout this thesis the term forest also applies to woodlands6. There are different types
of forest ecosystems at different latitudes namely boreal (CBD, 2008; WRI, 2009),
temperate (Ehrlich, 1996) and tropical forests (Foley et al., 2007; Nagendra, 2007), each
type of forest ecosystem having specific problems to face (WRI, 2009). Forests are
diverse, complex and dynamic systems, some being pristine while others are under several
human induced pressures (Powers, 1999; Spiecker, 2003). Tropical forests have been
suffering deforestation (Nagendra, 2007; Strassburg et al., 2009) whilst the area of

6
land under stands of trees with a canopy cover of at least 20% (or having the potential to achieve this),
including integral open space, and including felled areas. Definition by Forestry Commission UK
20
temperate and boreal forests has been increasing (Green et al., 2005; Mather and Needle,
1998; Nabuurs et al., 2001; Spiecker, 2003). The diversity in forest characteristics and type
of threats suffered gives an indication of the different meanings that SFM may have
depending upon the various socio-economic and ecological settings (Powers, 1999; Sayer
and Maginnis, 2005).

Forests are a type of ecosystem which is very difficult to manage (Spiecker, 2003;
Stengera et al., 2009). As was previously explained there are a multitude of interests for
management related to the provision of goods and services such as timber, water quality
and quantity or biodiversity to cite only a few (Sayer and Maginnis, 2005). Some
management practices aimed at enhancing yield of certain forest products are incompatible
with the provision of other goods at a local scale but might be compatible when placed
within the landscape scale (Stevens and Montegomerey, 2002). For example, a single tree
might provide an array of services useful for humans but if the timber is harvested the tree
ceases to provide other functions that it would have if it was not felled.

In addition, there is a huge diversity of types of forest ecosystems (Ehrlich, 1996; WRI,
2009). Different tree species and forest ecosystems have more or less capacity to act as a
carbon sink helping to mitigate climate change (Lexer et al., 2000; Powers, 1999). Not all
of the carbon fixed by unit area/time (the gross primary production GPP) is converted to
plant biomass instead; roughly 50-70 percent of fixed carbon is lost in respiration of
foliage and other woody tissues. That which remains is called net primary productivity
(NPP) a rate expressed as biomass production/unit/area (Powers, 1999). Forests produce
the greatest rates of NPP of any vegetation unit (Powers, 1999). At the same time, the
forest sector accounts for around 17 % of global green house gas emissions due to
deforestation and forest degradation (Hector and Bagchi, 2007; Strassburg et al., 2009). If
there is little doubt that the halting of deforestation and the replanting of large areas would
absorb CO2 while trees mature say for 40-60 years, it is also well known that it will only
“buy time” for developing other solutions to deal with climate change issues (Boyle, 2004;
Hector and Bagchi, 2007). Other problems arising from forests are: 1) intensive productive
systems may threaten the ecology of an area (Ehrlich, 1996), 2) problems arising from
invasive tree species and through forest fires may threaten sustainability of entire
landscapes (Cash et al., 2003; Clemente et al., 2005) .

There are a multitude of aspects that we as individuals, the forestry managers, researchers
and decision makers need to be aware of when dealing with forest management. Forests
have both the capacity to provide conditions for more sustainable landscapes but can also
be a threat to landscape sustainability.
21
At present, two main views exist to address forests‟ contribution to sustainability. These
are the ecosystems approach (EA) and sustainable forestry management (SFM). The
ecosystems approach (EA) was defined by the Convention on Biological Diversity as
follows and generally applies to all natural resources:

“a strategy for the integrated management of land, water and living resources that
promotes conservation and sustainable use in an equitable way. An ecosystems approach
is based on the application of appropriate scientific methodologies focussed on levels of
biological organisation, which encompasses the essential structures, processes, functions
and interactions among organisms and their environment. It recognises that humans, with
their cultural diversity, are an integral component of many ecosystems” (Sayer and
Maginnis, 2005:3).

The European strategy to promote SFM is implemented by the Ministerial Conference on

the Protection of Forests in Europe (MCPFE) which defined the concept as:

“the stewardship and use of forests and forests lands in a way, and at rate, that maintains
their biodiversity, productivity, regeneration capacity, vitality and their potential to fulfil,
now and in the future, relevant ecological economic and social functions, at local, national
and global levels, and that does not cause damage to other ecosystems” (Sayer and
Maginnis, 2005:2).

Although evolving separately, both concepts aim at promoting conservation and

management practices which are environmentally, socially and economically sustainable
and which generate and maintain benefits for both present and future generations (FAO,
2003; Sayer and Maginnis, 2005). Both concepts are guided by a set of principles. The
conceptual differences between the two sets of principles stem from different starting
points: production forests and forest management in SFM while EA focuses on
conservation ecology. Despite such conceptual differences a review from FAO (2003)
provides evidence for a full integration of the two concepts highlighting the need to
provide support for the actual implementation of both.

SFM has been addressed internationally, nationally and locally thus making it possible to
distinguish a hierarchy of institutions dealing with forests (Box 1.1). This hierarchy reflects
the different levels at which decisions relating to forest use and management are taken. At
the lowest level there are forest landowners, farmers, hunters, livestock grazers
representing different individual interests (Grimble and Chan, 1995). At intermediate
institutional levels there are the communal or municipal, state or provincial interests and
the national stakeholders that through the formal forestry offices deal with the
22
implementation of SFM (Sayer and Maginnis, 2005). At the international level different
institutions deal with sustainable forestry management. The United Nations (UN), in
addition to assuring the adoption of the statement of forest principles also, through its
Commission on Sustainable Development (UNCSD), established the Intergovernmental
Panel on Forests (IPF). The Food and Agriculture Organisation (FAO) is currently the lead
agency for forests in the UN. At the European level the European Forestry Commission
(EFC) is one of FAO‟s six regional forestry commissions.

One of the major problems in developing strategies for SFM is to precisely coordinate such
a diverse range of institutional scales (CBD, 2008; Sayer and Maginnis, 2005). Policies
often occur for, and affect, different scales, e.g. from international policies, to national
policies and finally to local policies that influence individual communities directly (Box
1.1). What is often missing is consistency among these various levels of policy, usually
because of the lack of direct connection among government departments, or between levels
of government (CBD, 2008).

An example of a cross-scale mechanism linked with SFM is that of forest certification.

Forest certification shares the aim of promoting SFM with another tool, namely Criteria
and Indicators (C&I) for SFM (Rametsteiner et al. 2003). Sets of C&I describe and
monitor status and trends in forests and forest management in parallel, so providing an
essential reference basis for forest certification (Olivier et al., 2000). Forest certification is
a seal of approval for forestry operations that implement good management practices. All
certification schemes are made up of three elements: standard, certification, and
accreditation. Box 1.5 defines these concepts.

Forest certification meets different interests. For industry and trade, it is an instrument for
environmental marketing and market access. For buyers and consumers, it provides
information on the impacts of products they purchase. For forest owners and managers, it
is a tool for market advantage. For governments, it is a means to introduce good
management practices and value the forest‟s resources (Rametsteiner et al. 2003).

23
Box 1.5. Forest certification

Standard defines the level of forest management practice that must be achieved. There are two
different types of standards: system standards and performance standards. System standards specify the
management systems that must be in place within an organisation to ensure it is managing quality and
environmental and social performance consistently. Performance standards specify the level of
performance or results that must be achieved in a forest. Summarizing, performance and system
standards deliver totally different outcomes and cannot be considered equivalent (WWF, 2000).

Certification is the process of establishing whether or not a standard has been met.

Accreditation is the mechanism for ensuring that the organisations that undertake certification (known
as certifiers or certification bodies) are competent and can produce credible results. Examples of
certifiers arise from several parts of the world namely from Canada, Canadian Standard‟s Association
(CSA); Australia, Australian Forestry Standard (AFS); Brazil, Sistema Brazileiro de Certificação
Florestal (CERFLOR) and Spain, Certificación Forestal (Certfor).

Source Sayer and Magginnis (2005)

With respect to certification goals, the Montreal process (WWW.mpci.org) [accessed 16

March 2009] was established to develop and implement internationally agreed upon
Criteria and Indicators (C&I) for the conservation and sustainable management of
temperate and boreal forests. With the same purpose, this time to represent the tropical
forests, the International Tropical Timber Organization (ITTO) established guidelines for
sustainable management for the forests in the tropics. Other groups, the Forest
Stewardship Council (FSC) whose members comprise non-governmental organisations
(NGO) and the private sector, have also developed principles and criteria for forest
management for a range of spatial scales (Higman et al., 1999).

It is clear in the literature that SFM is still an evolving concept. Siry et al. (2005)
concluded that the scientific community and forest managers must be “cautiously
optimistic” about SFM as defined by the Montreal Process, although much remains
unknown and much remains to be done. Wang (2004) also discuss the difficulty in
applying the concept on the ground. Debating the same issue, Higman et al. (1999) see
SFM as the forests contribution to SD but recognise the need to agree upon more precise
ways to implement it. Also Pearce (2001) notes that in some cases unsustainable forestry
management is more profitable than the gain that the certification process and SFM may
deliver.

Rather than enumerate several lists of C&I for different spatial scales it is important 1) to
coordinate efforts across a range of scales in order to attempt to implement SFM (CBD,
2008), 2) to recognise that there are numerous circumstances under which regional
development takes place and this is a precondition for understanding the multifaceted
nature of regional development in the forest sector (Munda 2005), 3) to explore the types

24
of multiple functions that post-modern societies are demanding from forests in order to
implement multifunctional forest management that meets the needs of the present and
future generations.

As is implicit in the two previous definitions (EA and SFM), forest management
institutions have recently shifted their management focus from sustaining yields to
sustaining ecosystems (Sayer and Maginnis, 2005). Accompanying such a shift, sustained
yield forestry, sustainable forestry, and sustainable forestry management represent a
progression of basic forest management concepts that demonstrates the recent trend to
greatly increase the range of goods and services for which sustainability is sought (Sayer
et al. 2005: 2).

One way of obtaining a multitude of goods and services from forests is to engage with a
broader group of stakeholders (Baskent and Yolasigmaz, 2000). Commercial interests,
environmentalists, recreationists and scientists should create constructive and sincere
partnerships for management decisions. This requires managing for a multiplicity of goals
and functions, so multifunctional forestry is of utmost importance in sustaining a range of
landscape types (Baskent and Yolasigmaz, 1999).

1.2.2. Multifunctionality

Multifunctionality implies the delivery of multiple functions from a single parcel of land
(Hagedorn, 2007; Mander et al., 2007; Zander et al., 2005). Multifunctionality is also often
interpreted as a jointness of production in which outcomes of the physical process of
production are considered as either positive or negative externalities (OECD, 2002, 2006).
In other words, multifunctionality implies a situation in which the effects of the delivery of
multiple functions or joint production is in accordance with the objectives of citizens and
politicians, such as having a beautiful landscape alongside agricultural production, and
other cases where joint production has a negative impact like nitrate pollution of ground
water (Mander et al., 2007). A concise summary of different interpretations of the concept
of multifunctionality was developed by Hagedorn (2007). Some interpretations focus on
the positive views of multifunctionality addressing, for example, the role of policies in
preserving the multifunctional character of sectors such as agriculture (Firmino, 1999).
Other interpretations use analytical tools to explore topics such as externalities, jointeness
of production in which “bads and goods” are accounted for (Mander et al., 2007). In
addition, other approaches address the issues of (re) creating and implementing
multifunctionality on the grounds that some aspects of multifunctionality from the past
might not viable in the present (Matthews and Selman, 2006). Multifunctionality can be

25
analysed both at the landscape scale as well as from activities such as forestry and
agriculture (Mander et al., 2007).

1.2.2.1. Multifunctionality: the forest scale

Multifunctionality is used to characterize the activities in the primary production sector

(Hagedorn, 2007). Sections 1.2.1.1 to 1.2.1.4 describe different facets of forest
management. The primary production sectors such as agriculture, forestry, horticulture
and related land dependent activities are considered as having a primary or main function
(production) and related joint production which typically include a mix of material and
non-tangible goods as well as a mix of private and public goods (externalities) (Lexer and
Brooks, 2005). Within the primary sector path (multifunctional) agriculture has received
most attention whilst the forestry sector is barely referred to but has a longer tradition of
multifunctional thinking (through terms such as multiple purpose or multiple use forestry).
Agriculture functions may be divided into primary and secondary (production and
externalities), whereas in forestry three functions (that are related to different management
systems) are normally recognised, namely production, recreation and protection (FAO,
2005; Lexer and Brooks, 2005). These three types of function are described in Box 1.6.

Box 1.6. Major functions of forest ecosystems

Production forest land designated/used for production and extraction of forest goods,
including both wood and non wood forest products.
Protection forest land designated/used for protection of soil and water also assuring
other protective functions such as biodiversity
Recreation forest land designated/used for the provision of social services including
recreation, tourism, education and or conservation of cultural spiritual sites.

Source: FAO (2005) Global forest resources assessment update: terms and definitions

Consequently, it can be said that in the primary sector path the term multifunctionality is
relatively well defined and accepted (Vejre et al., 2007). In both agriculture and forestry
activities the multifunctional character can be obtained in three different spatial ways: 1)
by pursuing different goals in a mixture of separate plots of land, 2) by pursuing different
goals on the same parcel of land, but sequentially in time, or 3) by integrating from the
beginning and coordinating the different goals to accomplish them simultaneously (de
Blust and Olmen, 2000). The first way can be defined as spatial multifunctionality because
different spatial units (plots) have clearly defined management goals. In spatial
multifunctionality each piece of land has one function, but when zooming in and out to
generate a mosaic the landscape appears to be more or less multifunctional. By contrast, in
26
the second and third variants, different goals are attained in the same spatial unit
(successively in the second and simultaneously in the third). This type of multifunctional
land use can be defined as integrated multifunctionality (de Blust and Olmen, 2000).
Figure 1.1 shows the two types of multifunctionality at a forest scale.

Stand scale/ plot of land (e.g.50 ha)

Pr /Pt/Rc Integrated multifunctionality

Spatial multifunctionality (in

Pr Rc Pt forestry also called multiple
use by adjency)

Pr-Production, Pt-Protection, Rc-Recreation see Box 1.6, adapted from de Blust and
Olmen (2000)
Figure 1.1. Multifunctionality at forest scale

As was previously stated in Section 1.1, the role of forestry is changing to include not only
productive, but also consumptive (recreation) and protective functions. Achieving this will
require a new coordination of forest functions in post-industrial landscapes that certainly
needs to be carefully addressed (Nabuurs et al., 2001; Niskanen and Lin, 2001; Slee,
2007a). This “new” holistic approach for forest management through landscapes tends to
not physically and permanently divide the land base for certain uses, instead, it aims to
manage the forest ecosystems as a whole to provide multiple functions simultaneously
(Baskent and Yolasigmaz, 2000; Lexer and Brooks, 2005). The next section explores the
goods and services provided by forests.

1.2.2.2. Goods and services provided by forests

The three functions of forests described in Box 1.6 (related to three different management
systems) provide a variety of goods and services. Goods and services provided by forests
can be included in a broader category of goods and services provided by the environment
(de Groot et al., 2002; Slee, 2007a; Turner and Daily, 2008). Despite the large body of
literature on ecosystem (or landscape) functions, goods and services, there is still not a
clear consensus on the final definitions of these concepts (de Groot and Hein, 2007);
especially problematic is the distinction between function and service (Fisher et al., 2004).
The last report of the Millenium Ecosystem Assessement in 2005 made an attempt to bring
order to the many definitions of “functions”, “goods” and “services”. Agreement was

27
reached to define services as “the benefits people derive from ecosystems”, and in order to
avoid lengthy texts it was decided to use the term “services” for both goods and services as
well as the underlying functional processes and components of the ecosystems providing
them (de Groot and Hein, 2007). Many authors however, have highlighted a principal
difference between the term function and service. For example de Groot et al (2002:394)
defined function as “...the capacity of ecosystems to provide goods and services that satisfy
human needs either directly or indirectly”. Functions are seen as the actual (functional)
processes and components in ecosystems and landscapes that provide, directly or
indirectly, goods and services which benefit human welfare (de Groot and Hein, 2007).

The term forest functions implies a normative–ontological approach to the analysis of

relationships between society and the environment, in which natural resources are seen as
entities providing „functions‟ for society. The concept of „forest functions‟ also implies
that forest land-use management is a task to be managed and planned (forests can be
managed for production, protection and recreation). This is contested by some scholars
(Janse and Ottitsch, 2005).

Acknowledging that there is not one classification scheme that will be adequate for the
multitude of contexts (e.g. decision making or research) in which ecosystem service may
be important, Fisher et al (2004) argued that any attempt at classifying ecosystem services
should be based on both the characteristics of interest and a decision context. Decision
contexts addressed by Fisher et al (2004) were education and understanding, valuation,
land management and distribution and equity in human welfare. The land management
context defined by Fisher et al (2004) was considered useful for the purpose of this
research and a discussion of this follows.

Describing relationships between service production and where the benefits are realized (as
explored by Fisher et al 2004) is important in the forestry context. Box 1.7 and Figure 1.2
show that the manner in which forest services are provided (P) and benefits (B) delivered
may occur in at least three different ways: 1) in situ, 2) omni-directional and 3) directional.

28
Box 1.7. Places where services are provided (P) and benefits delivered (B)

1. In situ. The services are provided and the benefits are realized in the same location
(e.g. soil formation, provision of raw materials).
2. Omni-directional. The services are provided in one location, but benefit the
surrounding landscape without directional bias (e.g. carbon sequestration, pollination)
3. and 4. Directional. The service provision benefits a specific location due to the
flow direction. In 3 down slope units benefit from services provided in uphill areas,
for example water. In 4 the service provision unit could be coastal wetlands (or
forests) providing storm and flood protection to a coastline.

Source: Fisher et al (2004) p. 12

Figure 1.2. Places where services are provided (P) and benefits delivered (B)

As previously discussed to the most common functions attributed to forestry are those of
production, protection and recreation (Box 1.6). Generally, production comprises both
timber and non timber products. The terms Non-Wood Forest Products and Services
(NWFPS) or Non Timber forest products (NTFP) are also used when talking about the
broad scope of functions, besides timber production, that forests fulfil.

At present, it is increasingly recognised that a variety of environmental services provided

by forests are very important and thus have an economic value (Pearce, 2001). For
example, in the well known paper by Costanza et al. (1997) (and widely criticized see
Balmford et al, 2002) the economic value of forests represented 30 % of the total economic

29
value of the world ecosystems (Balmford et al., 2002; Costanza et al., 1997; Slee, 2007a).
According to Pearce (2001:284) “all ecological functions of forests are also economic
functions” however they may either represent a gain for society (biodiversity, timber and
non timber products) or be an expense (e.g. prevention of wild fire, combat invasive tree
species). Furthermore, many important forest functions, although valued, have no markets
and hence no apparent economic value (Pearce, 2001). This is one of the causes of the loss
of forest area that occurs mainly in the tropics (Balmford et al., 2002; Nagendra, 2007).

Despite this, with either emergent or indirect markets, there are studies reporting the
economic value of services provided by forests. For example, Bateman et al. (2005)
calculated the economic value of recreation in woodlands and forests for England. The
problems of transferring values across places are well known but these values reveal the
economic importance of forests for recreation activities.

One contemporary expectation of humans regarding forests is to use their characteristics to

overcome global warming threats (Karjalainen et al., 2003). This can occur either by using
forests ability to sequester carbon in living wood and soils (Bateman and Lovett, 2000) or
their capacity to produce “green” energy through different processes that transform forest
biomass (Boyle, 2004). However, there is a strong demand for scientific estimates of the
current and potential contribution of forests in both carbon sequestration (wood and forest
soil) and forest biomass as a renewable source of energy (Karjalainen et al., 2003; Schmid
et al., 2006). Concerning carbon sequestration, Schmid et al. (2006) investigated the effect
of forest management on future carbon pools and fluxes. This study shows that different
forest management strategies alter the ability of forests to act as a carbon sink or source
(Schmid et al., 2006). Regarding forestry biomass, recent studies indicate that forestry
practices such as whole tree harvesting (log+ wood residuals in order to produce renewable
energy) will cause net losses of nitrogen in some forest areas which means that forestry
will not be sustainable unless nutrients are added through compensatory fertilization
(Akselsson et al., 2007).

In addition to timber and non timber products, recreation, carbon sequestration and forest
biomass, forests also provide other environmental functions such as watershed protection,
sewage treatment and noise reduction. Table 1.8 shows different forest goods/services
according to the places in which services are delivered and benefits occur. The problem is
that whether they are valued or not, most of the functions from forests (that become
services) do not have effective markets (Pearce, 2001; Slee, 2007a).

30
There are however, some financial mechanisms through which stakeholders downstream
can provide payments for afforestation projects to stakeholders located upstream in the
catchment (Johnson et al., 2002). A review of financial mechanisms for developing
markets for water services from forests is presented by Johnson et al. (2002) and Sabatier
et al. (2005). This topic is further discussed in Chapter 4.

Some of the goods and services provided by forests described in Table 1.8 might be
generated either by dominant or multifunctional use. The next section describes
multifunctional and dominant use types of forestry.

31
Table 1.8. Goods and services provided by forests

P/B Good/ Description Sources

Service

In situ Timber Timber is a product which needs no detailed description. Prices of (Bateman
timber vary according with tree species as well as the timber et al.,
characteristics and commercial circuit. Bateman et al (2005:111) 2005)
assess both the social and private value of timber production in the
UK.

Non timber There are a multitude of non-timber products that forests may (Janse and
products provide. Examples include mushrooms, medicinal plants, cork, and Ottitsch,
rubber. These may be traded in a variety of ways such as distant 2005)
markets or be part of local development strategies. The income
generated by selling NTP varies from product to product and the
length of the trade circuit.

Forest Biomass energy (which includes forestry biomass) is an emerging (DTI,

biomass industry which does not yet have established supply chains or 2007)
quality standards. As a result biomass prices may be quite variable
year on year reflecting production/availability, and between (Boyle,
localities reflecting differing supply demand balances. There are 2004)
only limited markets for forest biomass so the normal
recommendations are : 1) small scale production is mainly relevant
to local utilisation, for example for heat supply to buildings and
dwellings within an estate. 2) the price for medium to large volumes
of wood chip will tend to be set by the cost of collection, processing
and transport plus a small margin. A survey of published reports
found supply price estimates ranging from £ 0.77/GJ to £3.2/GJ.

Recreation The recreational aspects are perhaps the most valued ecosystems (Brainard
and services for urban dwellers. There are different ways in each et al.,
cultural recreation may generate income. In the rural areas across Europe for 2001)
values example forest owners in the Netherlands focus on offering small-
scale, nature-based facilities for (short-stay) recreationists, whereas (Bateman
for Norwegian forest owners it is more promising to focus on et al.,
offering wilderness-experience 'all-in package-deals' to tourists. 2005)

Air Vegetation reduces air pollution but to what level seems to depend (Bolund
filtering on the local situation. The reduction is caused by vegetation and
filtering pollution from the air. Filtering capacity increases with leaf Hunhamm
area so is higher for trees than bush and grassland. Because of the ar, 1999)
larger total surfaces of needles, coniferous trees have a larger filter
capacity than the trees with deciduous leaves. This capacity is also
grater because the needles are also shed during the winter. But
coniferous trees are sensitive to air pollution and deciduous trees are
better to absorb gases. Thick vegetation may simple cause
turbulence in the air while thinner cover may let the air through and
filter it. 1 ha of mixed forest removes 15 tonnes of particles per
year. Pure spruce forest may filter 2 or 3 times as much.

Micro- Local climate and even weather is affected by places such as cities. (Bolund
climate This phenomenon, called urban heat island effect, is caused by the and
regulation, large area of heat absorbing surfaces, in combination with amounts Hunhamm
at street of energy use in cities. All natural systems help to reduce these ar, 1999)
and city differences. Trees can lower summer temperatures markedly by
level decrease energy use for heating and air-conditioning when shading
houses in the summer and reducing wind speed in winter. A single
large tree can transpire 450 l of water per day. An increase of 10%
in tree cover, or planting three trees per building lot could reduce
the energy expenses by 50-90 US dollars per dwelling unit per year

32
 Good/ Description Sources
Service

Noise Vegetation contributes to a decrease in the propagation noise but at (Bolund

reduction what level is uncertain. Different studies achieve very different and
results. Though of value there are still no markets for these services Hunhamm
provided by forests nor were market values found. ar, 1999)

Sewage Taking care of sewage costs, mainly in cities large amounts of (Bolund
treatment money, and the nutrients that are still released contribute to and
eutrophication of the surrounding water ecosystems. Some studies Hunhamm
shown that wetlands and forests can significantly reduce the costs of ar, 1999)
sewage treatments. Though of value there are still no markets for
these services provided by forests nor were market values found.

Directio Rain In vegetated areas only 5-15% of the rain water runs off the ground, (Bolund
nal water with the rest evaporating or infiltrating the ground. In vegetation and
drainage free areas such as cities about 60% of the rainwater is instead Hunhamm
disposed of through storm water drains. Valuation of this service ar, 1999)
depends upon the local situation.

Flood Forested slopes provide both water retention and avoid landslides at (Fisher et
protection areas located downhill. Forests located in coastal areas also provide al., 2004;
storm and flood protection to a coastline. Though of value there are Johnson et
still no markets for these services. There are however arrangements al., 2002)
in place in order to manage this service with the whole catchment
vision

Whole In a area of a single watershed the biophysical relationships between (Johnson

catchment forests, people and water are highly variable. There are however, et al.,
manageme basic relationships that reflect the importance of forests in the whole 2002;
nt catchment management: 1) Forests slow the rate of runoff in a Sabatier et
watershed, 2) Forests reduce soil erosion and sedimentation of al., 2005)
waterways, 3) Forest soils filter contaminants and influence water
chemistry.

Omni Carbon The global carbon market established in 2001 as part of the Kyoto (Basu,
direction sequestrati protocol allow the trade of carbon credits e.g. Clean Development 2009;
al on Mechanism CDM which are systems by which participating Karjalaine
countries or institutions can meet some of their greenhouse gas n et al.,
reductions by buying certified carbon credits. Prices vary. In India 2003;
were reported at $65 per year for each hectare committed to grow Schmid et
they forests for four years rather than logging every year. European al., 2006)
Commission initiated the emission trade schemes in 2005 see Box
1.4). There is a suite of techniques for predicting carbon pools and
fluxes. All the models used for such purposes are situated
somewhere on the gradient between empirical based models and
process models. The different models may result in different carbon
estimates at the ecosystem scale.

Pollination The presence of trees and forests provides habitat for pollinators (de Groot
such as bees. Pollination function is very important in assuring food and Hein,
security. 2007; de
Groot et
al., 2002)

33
1.2.2.3. Multifunctional forestry vs. dominant use

Management of forests has evolved from dominant use to multiple use, from product
output focus to an ecosystem health focus, from emphasis on one use or product to the
joint production of multiple products and services (Stevens and Montegomerey, 2002).
Timber use production is the most prominent of the dominant uses though the
simultaneous use of the same piece of land for several purposes allows the “simultaneous
or joint exploitation” of different products from the same plot of land. However, the
consumption or use of one forest product or service may have an effect on other products
and functions. For example, clear-cutting a mix aged, mixed species forest and replanting
to a single species forest reduces biodiversity. Multi resource management is a challenge
because

“Simultaneous use of the same piece of land for several purposes is often difficult since
many uses compete with as well as supplement each other. Maximum production of timber
interferes with maximum production of wildlife. Full utilization of forage reduces the yield
of the wood. Heavy cutting may make the forests less effective as a regulator of runoff and
certainly impairs its value for recreation. Complete preservation of natural conditions for
the benefit of the water supply or the nature lover puts a stop to all industrial use” Dana
(1943) cited by Stevens and Montegomerey (2002:1).

Research on tradeoffs and complementarity of production in the multiresource forest

environment helps guide these choices. According to Stevens and Montegomerey (2002)
multiresource research in forestry has two major components namely valuation and
production possibilities. The former refers to public preferences and social values guiding
the desirability of increasing one forest use even when it implies decreasing another, the
latter refers to the productive capacity of the land and the compatibility of one forest use
with another. Compatibility research tries to provide guidance on the cost of increasing one
forest use at the expense of other forest uses (Stevens and Montegomerey, 2002).

One of the first landmarks proclaiming the shift from dominant to multiple use forestry
occurred in the US through the Multiple-Use and Sustained Yield Act (MUSY) of 1960.
Land use planning under the MUSY Act aimed at coordinating potentially conflicting uses
rather than zoning for single uses, however often the management result was viewed as
adjacent, single resource land use allocations. Although the call was for integration of uses
the feeling prevailed that forest management had often followed a zoning model of many
dominant uses named as “multiple use by adjacency” (Stevens and Montegomerey, 2002).

34
In this respect, scale is of extreme importance because what may be multiple uses at the
forest level may actually be dominant use at the stand level. Moreover, uses can be
considered incompatible when investigated at smaller scales and compatible when broader
scales are considered (Stevens and Montegomerey, 2002). Forest uses are more likely to be
compatible at regional or forest scale than at the smaller stand or management unit scale.

If there are a multitude of studies addressing valuation of forest goods and services there
are fewer studies addressing compatibility of forests at the landscape scale (de Groot,
2006). According to Stevens and Montegomery (2002) the majority of the compatibility
studies of forests have been carried out at stand level. Above all, multiple vs dominant use
are clearly different in ways to look at forestry. Both have strengths and/or weaknesses and
are more or less appropriate for different environmental and socio-economic dynamics. At
the present, multifunctional forest management has been increasingly recognised as crucial
in moving towards SD through landscapes but there is a need to further investigate the
types of multifunctionality able to do so (Pinto-Correia and Breman, 2008).

1.2.2.4. Good forestry practices

The types of functions that forests are able to provide (production, protection, recreation)
deliver different goods and services such as production of timber and biodiversity (see
Table 1.9). As a consequence, depending on the management system undertaken, forests
will provide more or less timber as well be more or less appropriate for biodiversity
requirements.

Table 1.9. Forest functions and management strategies

Function Production Recreation Protection

Management approach Plantation Nature based integrative Nature conservation
Specific management Focus on timber Recreation, flexible Strict forest reserves
goals production and direct wood production and providing uneven natural
economic outcome nature protection structures and processes
Production of timber +++++ ++++ +
Landscape beauty ++ +++ +++++
Recreation ++ ++++ ++
Biodiversity + +++ +++++

Comparison of different management approaches with an indication of their respective fulfilment

of specific management goals. The goal fulfilment is subjectively scored on a scale from 1 to 5
plusses, where “+” = low goal fulfilment and “+++++”= high goal fulfilment. The table shows
some basic principles and general features of the three management approaches to forestry.
Source: Adapted from Sayer and Maginnis (2005:62)

35
Although different management systems have specific “good practices” for example, a
good practice in a forest managed for recreation is to vary the stand composition by putting
together broadleaves and coniferous forests and also varying the stand ages (different
levels of canopy), some general rules for good forest management practices may be
defined. One of the “universal” good practice rules in order to maintain the natural capital
of an area is not to surpass the yield capacity (Costanza and Daly, 1992). Another is that
of incentivising Continuous Cover Forestry (CCF) which despite not being a new idea in
forest management has been of renewed interest for the potential it has to meet
sustainability requirements (Pommerening and Murphy, 2004). One of the reasons for
supporting CCF is the need to overcome the recent trends of fragmentation in forest areas
(Fahrig, 2003). Another matter of concern in forest management at present is to overcome
the trends towards simplification of stand structures caused mostly by even aged
plantations in the past. Thus, promoting mixed aged class stands (see Table 1.10) is needed
and is an important feature of SFM. Another universal rule is to manage forests in such a
way that the sanitary conditions of forests are assured (Ehrlich, 1996). Avoiding the spread
of diseases as well as removing litter from forest plots is very important in SFM. In
forestry, the notion of strategic (or long range) planning is widely known with origins that
can be traced back to centuries ago. The most distinctive characteristic of forestry is the
long length of the production process (known as the rotation period).

In some temperate forests of North America and in Mediterranean countries as well as in

Australia, the risk of wild fires is an important matter of concern. Fire causes important
changes in environmental conditions for plant growth and establishment both due to the
destruction of above ground biomass as well as direct effects on soil physical properties
and biogeochemical processes (Clemente et al., 2005). In order to overcome the severe
effects of wild fires the use of controlled fires has been increasingly recognised to be a
“good management practice”. There are studies reporting the importance of controlled fires
breaks in continuous forests to avoid a rapid spread of fire (Fernandes, 2001). Other
techniques for fire management are variation of stand structures because in an uneven
stand fire progression is slower due to different wood densities.

36
Table 1.10. Good forest management practices

Good forest management practice Description and sources

Continuous cover forestry (when Includes those sylvicultural systems which involve
possible) continuous and uninterrupted maintenance of forest
cover avoiding clearcutting.

Awareness about negative effects Contiguous high forests such as beech can have negative
of CCF effects on tree species diversity because less competitive
species tend to be extinguished compared to middle or
low forests7

Variety of stand structures (mixed Diversification from monospecies coniferous plantations

age classes and tree species)
reduces biotic, abiotic and economic risks (e.g. diseases
and fire cannot spread as easily as in pure stands). Mixed
forests provide a wider range of size classes and timber
products allowing flexible and rapid response to market
conditions without decreasing total volume production

Using protective silvicultural Establishment of mixed stands sometimes implies to

techniques such as nurse crops
have shelter trees or nurse crops. The procedure is to
establish a nurse crop of a pioneer species by planting or
seeding at a comparatively wide spacing. Afterwards the
target species are introduced being protected against
extremes of weather from the pioneer species.

Attention to site limitations Tree species and provenance choice should be dependent
on site conditions. Native tree species and broadleaves
should be favoured.

Conservation of old trees, Retaining a certain amount of lying and standing

deadwood and protection of rare deadwood in each forest stand is recommended both for
species biodiversity and amenity reasons (though some studies
show that the presence of deadwood decreases
recreational value)

Spatial arrangement maters Tackle fragmentation issues by enhancing forest

networks. Establishment of forest margins as transition
zones between the open landscape and forests.

Source. Adapted from (Pommerening and Murphy, 2004)

This section has indicated that forests may provide an array of goods and services that
extend beyond the place in which forests are located. There are a multitude of management
strategies that can differentially deliver goods and services from forests. Multifunctional
forest management was described as of utmost importance. The next section explores the
concept of multifunctionality from a landscape perspective.

7
Low forests are forests entirely managed on a short rotation basis by coppicing (coppice system) while
middle forests (coppice with standards) are result of coppice though allowing some individual tree to full
growth Pommering and Murphy (2004)
37
1.2.3. Multifunctional landscapes

In the landscape setting (or path), multifunctionality is used to characterize the landscape
per se (Vejre et al., 2007). In contrast to the primary sector, the actors within the landscape
sciences have not reached a common agreement on the terms function and hence
multifunctionality (Fry, 2001; Vejre et al., 2007). Compared to agriculture and forestry,
where functions such as production, protection and recreation are straightforward to reach
agreement upon, functions of landscapes are not by definition grouped, rather they are
difficult to realise due to the complex character of functionality in landscape ecosystems.
Also, as was previously pointed out, at the landscape scale the full spectrum of ecosystem
processes or functions can be realised, whereas at the level of one field fewer functions
(e.g. productivity) appear (Mander et al., 2007).

Furthermore, the functioning of ecosystems depends upon processes that take place over a
range of ecosystems and institutional scales (Box 1.1). Ecological organization at a specific
scale is determined mainly by interactions between species and processes operating within
that scale (Peterson et al., 2008). Competitive interactions are strongest among species that
have similar functions and operate at similar scale. These interactions encourage functional
diversity within a scale and the distribution of ecological functions across scales (Peterson
et al., 2008). Thus embracing the complexity of multifunctionality at landscape level is an
enormous challenge.

The Millennium Ecosystem Assessment initiative (MEA 2003, 2005) has nevertheless
distinguished four groups of landscape functions namely provisioning, regulating, cultural
and supporting. Based on a modified version of the MEA (2005) (supporting was replaced
by habitat function because the paper focussed on economic valuation and there was the
risk of double counting some services) de Groot and Hein (2007) analysed the types of
landscape functions and their implications for the ecological and institutional scales shown
in Box 1.1. As can be seen from the description of the different landscape functions in
Table 1.11, the benefits of the resources may accumulate to stakeholders at a range of
institutional scales.

Production functions are concerned with the possibility of “harvesting” products from
natural or semi-natural landscapes and this depends upon the availability of the resource.
To analyse the ecological impacts of the resource use, or the harvest that can be supported,
the appropriate scale of analysis is the level of the landscape supplying the service.
However, the benefits of the resource may accumulate to stakeholders at a range of
institutional scales. At local scale, residents, if present, are often an important actor. In

38
addition, there may be stakeholder interests at wider scales if the goods involved are
harvested, processed or consumed at larger spatial scales.

Regulation and habitat functions can be interpreted as ecological processes that have
(actual or potential) economic value because they may have an economic impact outside
the studied landscape and/or if they provide a direct benefit to people living in the area
(Table 1.11). For some regulation functions not only the scale is important but additionally
the position in the landscape plays a role; for example, the impact of the water buffering
capacity of forests will be noticed downstream in the same catchment (Box 1.7 Figure 1.2).

Cultural functions are also supplied at different ecological and institutional scales. As the
value attached to the cultural service depends on the background of the stakeholders
involved there may be very different perceptions of such values at different scales.

The multifunctionality of forestry and agriculture (Section 1.2.2) and the multifunctionality
of the landscape (Section 1.2.3) have land use as a common denominator. Land use is a
term that makes sense in describing the functionality of farming and forestry as well as the
landscape systems (Vejre et al., 2007). It is obvious that between agricultural or forestry
and landscape viewpoints there is the issue of scale. It is, however, likely that the types of
function at the forest scale will have effects on landscape multifunctionality. For example,
a large plot of forest primarily managed for production is likely to provide few
opportunities for scenic landscapes in the area it is located. In addition, combining forests
with surrounding land uses such as urban or agriculture will certainly create a multitude of
different functions at the landscape level.

As was previously discussed, multifunctionality has also to be seen in terms of negative

aspects, arising not necessarily only from the jointness of production. The multiple
functions occurring in a landscape also imply a linkage between a community and its
environment (landscape) and in some regions this is at peril (Antrop, 2006; Selman, 2006;
Vejre et al., 2007). As was discussed in Section 1.1, it is acknowledged that some protected
and cultural (Box 1.2) landscapes in post-industrial societies are not self- sustaining
because the links between landscapes, community and economy are no longer self
reinforcing (Mander et al., 2007; Selman, 2006). As a consequence there are
multifunctional landscapes in which the link with sustainability is loose (Hagedorn, 2007;
Mander et al., 2007). As there are different types of landscapes one of the factors varying
between them is the amount and type of forest ecosystems and, forests may be only a part
of the multiple ecosystems across landscapes. Nevertheless, it is important to investigate

39
the type of forestry that is likely to contribute to more sustainable landscapes (Pinto-
Correia and Breman, 2008).

Table 1.11. Description of landscape functions

Landscape Description
functions

Provision To analyse the ecological impacts of the resource use, or the harvest that can be
functions supported, the appropriate scale of analysis is the level of the landscape supplying
the service. However, the benefits of the resource may accumulate to stakeholders
at a range of institutional scales.

Regulation Regulation services are typically generated at a specific ecological scale but the
functions benefits may accrue to stakeholders at a range of institutional scales. For many
regulation services not only the scale but the position in the landscape plays a role-
for example, the impact of the water buffering capacity of forests will be noticed
only downstream in the same catchment. Stakeholders in a regulation service are
all people residing in or otherwise depending upon the area affected by the service.

Habitat Habitat functions refer in particular to the opportunity to maintain and protect
functions biological diversity offered by landscape-as an end in itself. Stakeholders of habitat
functions are all people residing in or otherwise depending upon the area affected
by the service.

Cultural and Cultural and amenity function are supplied by landscapes at different ecological
amenity scales, such as the presence of a monumental tree or a natural park. Stakeholders in
functions cultural services can vary from the individual to the global scale. Nature tourism
(most has become a major cultural service in western countries and has linked
common is stakeholders from the local to global scales. The value of the service may differ
recreation) e.g. local stakeholders may attach particular value to local heritage whereas
national/global stakeholders may have a particular interest in the conservation of
nature and biodiversity.

Based on de Groot and Hein (2007:30)

To summarise, Section 1.2 started by defining the concept of sustainable forestry

management, this aims at coordinating the ways in which forests may contribute to more
sustainable landscapes. In order to do so, SFM needs to be addressed at a multitude of
ecological and institutional scales (Box 1.1). Following this, the concept of
multifunctionality was explored at the forest scale (Section 1.2.2) as well as at the
landscape scale (Section 1.2.3). It was stressed that forests may provide a multitude of
functions that may lead to landscape sustainability and this is better achieved if
multifunctional forest management is put into practice. The next section presents the
possible ways to move forward by specifically addressing the tools and methods that might
help to inform SFM at different spatial scales.

40
1.3. Sustainable landscape planning and landscape governance as a way
to move forward
Sustainable planning seeks to link knowledge about sustainability with actions to achieve
it (Ahern, 2005). Sustainable planning therefore attempts to “implement” or
“operationalise” the principles of sustainability in planning (Kato and Ahern, 2008).

According to the European Landscape Convention, planning is “a forward-looking action

to enhance, restore or create landscapes” (Council of Europe, 2000). Planning is generally
associated with the integration within place or, in other words, a “place making”
instrument. As was previously highlighted (Section 1.1) a spatial planning approach is thus
a tool for planning for the “whole landscape”. A “whole landscape approach” was defined
by Dolman et al, (2001:306) as “a process of integrated planning across property
boundaries that optimizes the amount, location, configuration and management
of...landscape elements”. Management, on the other hand, means the routine tasks required
in order to achieve the planning goals.

Nevertheless, the mere act of establishing goals and adopting programs (in planning),
however of great support, will never result in more than just goals and programs unless
these are made fully operational and enforceable in order to achieve the stated goals
(Carlman, 2005). Landscape governance aims at implementing the stated planning goals by
working at the institutional level (Adger et al., 2004). It aims at delivering policies or
mechanisms (e.g. tax reduction) able to implement the stated planning goals on the ground
(Milligan and O'Riordan, 2007). Thus, governance for sustainability can be understood as
“the emergence and the implementation of innovative shared forms of planning and
managing of socio-spatial dynamics” (CEMAT, 2006:29). Landscape governance
furnishes an effective setting for the operation of area-based partnerships which can
intervene in a more integrated and place-sensitive manner (Selman, 2006).

It follows from the above that planning and governance for sustainability is an immense
challenge (Forman and Collinge, 1997; Nicholson-Cole and O'Riordan, 2009). One of the
roles of planning/governance systems is to navigate through multiple realities and
recognize uncertainty (i.e. to address fuzzy8 concepts and doctrines) in an attempt to
identify appropriate responses for a specific socio-economic and ecological (SESs)
context. These specific socio-ecological (SESs) contexts will create different forms of
agriculture and forestry which in turn will influence the multifunctional character of the
landscape (Sections 1.2.1 and 1.2.2). In these settings, establishing the linkage between the
concept of landscape with the concept of steered or planned sustainability is, if not utopia
8
The word fuzzy is used to refer to situations where there is lack of clarity in the basic concepts
41
(Antrop, 2006), at least, challenging. How can landscapes (seen as holistic) be planned?
Even more importantly is how can landscapes be planned for sustainability?

Sustainable landscape planning and landscape governance aim at coordinating the

processes occurring between man and nature (Naveh, 2007). Landscape planning occurs at
different planning levels covering different spatial scales (von Haaren and Ott, 2008). The
scale at which an analysis is undertaken significantly influences the problem definition,
methodological approaches used and ultimately the outcome of the analysis. Recognition
of the importance of scale in the context of environmental resources assessment and
management has grown considerably over the past decade (Cash et al., 2006; MEA, 2003).
For example, the Millenium Ecosystem Assessement (MEA) used a multiscale approach
that consisted of component assessments undertaken at multiple spatial scales ranging from
individual villages to the globe. In the MEA there were, however, different categories of
sub-global assessements (MEA, 2003, 2005) representing those different forms of relating
and cross scaling the different analyses undertaken. Multiscale assessements are both
resource and time intensive but they do provide a “complete insight” to the problems
addressed (MEA, 2003) as they allow a focus on both interactions within or across scales.
According to Cash et al (2006:2) “cross scale” means the interactions across different
scales.

There are also multiscale studies such as those by Herrmann and Osinki (1999) and Lopes-
Ridaura et al. (2005) in which different spatial levels were combined within a framework.
In such a multiscale analysis the highest level provides general ideas for the development
of planning measures at the lower scales of analysis (Lopes-Ridaura et al., 2005). As the
problem definition and therefore the research questions are likely to vary between spatial
scales, different methods are likely to be more or less appropriate for different scales of
analysis (Herrmann and Osinski, 1999). In general, quantitative approaches are used at
broader spatial scales because of their inherent analytical power. By contrast, at more
detailed spatial scales in-depth studies usually require qualitative approaches (Flick, 2002).
Qualitative approaches may be put into practice in a variety of forms such as focus groups
and interviews. Flick (2002) reviews the strength and weakness of different qualitative
methods.

Studies describing challenges, barriers, strategies and recommendations for the

implementation of sustainable landscapes are reviewed in Ahern (2005). There are also
studies reporting the way in which planning provided mechanisms through which
communities developed sustainability discourses (Hanna, 2005). The literature on
sustainable landscape planning refers to several components that landscape planning and
42
landscape governance with sustainability goals must address in order to move towards
more sustainable landscapes (von Haaren and Ott, 2008).

Table 1.12 summarizes some important principles for sustainable landscape planning and
landscape governance. As highlighted in Table 1.12, in order to plan for sustainability,
none of the disciplines of ecology, economics, and institutional theory, as construed at
present, can, in isolation, deal with the challenges at stake, instead transdisciplinary
approaches should be used as the guiding principle (Naveh, 2007). Transdisciplinarity is a
approach where integrative forms of research (comprising more than one research field)
interact with extra-scientific experience and practice (often involving stakeholders outside
science) in concrete problem-solving (Tress and Tress, 2001). This implies the use and the
integration of knowledge from research fields such as environmental sciences, social
sciences and economics together with practitioners and stakeholders into concrete problem
solving (Naveh, 2007)

Seeking to implement sustainable landscape planning also involves a wide range of

techniques tools and methods from research fields such as environmental and social
sciences. Disciplines such as landscape ecology study the composition and configuration of
landscape patches in order to characterise functions and processes operating in it (Dolman
et al., 2001; Forman, 1995; Hersperger and Forman, 2003; Naveh, 2007). One widely used
tool in landscape ecology is the calculation of landscape (or class) metrics by using
different types of software, of which the most common is the program FRAGSTATS. This
software computes a wide variety of landscape metrics for categorical map patterns
(McGarigal et al., 2002). The original software (version 2) was released in the public
domain during 1995. FRAGSTATS computes several statistics for each patch and class
(patch type) in the landscape and for the landscape as a whole. Class metrics represent the
spatial distribution and pattern within a landscape of a single patch type while, landscape
metrics represent the spatial pattern of the entire landscape mosaic, considering all patch
types simultaneously (McGarigal et al., 2002). The problem is, even though many of the
indices have counterparts at the class and landscape levels, their interpretations may be
somewhat different. Both at the class and landscape level, some of the metrics quantify
landscape composition, while others quantify landscape configuration. There are different
studies calculating metrics for forests (both class and landscape) in the study of ecological
process such as fragmentation (Fahrig, 2003; McGarigal et al., 2002). Other researchers
have highlighted the usefulness of landscape metrics in addressing sustainability
(Botequilha Leitao and Ahern, 2002) (Chapter 3 further explores this topic).

43
Social scientists have, and will keep adding, valuable knowledge in dealing with
stakeholder engagement throughout planning and governance processes (Milligan et al.,
2009). The use of participatory approaches in natural resource management is a well
known research field (Grimble and Chan, 1995). Based generally in participatory
approaches, environmental social sciences aim at producing understanding that can be
applied to the resolution of environmental conflicts by exploring institutional arrangements
and governance systems likely to promote sustainable management of natural resources
(Adger et al., 2004).

For Sustainable Forestry Management (SFM) the involvement of stakeholders is crucial

and various studies have examined stakeholders preferences for multiple forest values
(Kummar and Kant, 2007). As was previously discussed in Section 1.1 there are
differences in public preferences for forests across stakeholders groups (Lewis and
Sheppard, 2006). In recent years the use of visualization tools has been increasingly
recognised as of utmost importance and several studies have used these in order to
effectively engage the public in environmental management (Appleton and Lovett, 2003;
Appleton, 2003; Daniel and Meitner, 2001; Wang et al., 2006). There are also studies such
as that by Tahvanainen et al. (2001) which addressed public preferences for forests based
on both visual and verbal data. There is a huge body of literature studying the effects that
public preferences may have on the ecological functioning of ecosystems (Gobster et al.,
2007; Tahvanainen et al., 2001). In these types of studies either questionnaires including a
set of photos (photo-questionnaire) or interviews and focus group meetings are widely used
(Bogaert, 2002; Winter, 2005) (Chapter 5 addresses this topic).

Economists have a crucial role in their work on ecosystem and landscape valuation
(DEFRA, 2006). An enormous variety of valuation methods have been developed and are
of utmost importance in informing policy makers worldwide (Balmford et al., 2002; de
Groot et al., 2002; Fisher et al., 2004). There are a variety of methods used to value
different goods and services provided by forests (Pearce, 2001; Slee, 2007a; Stengera et
al., 2009) and the value of different forest functions has been calculated in a range of
studies (Pearce, 2001).

As shown in Table 1.12, another important feature in planning for sustainability is to deal
with both diversity and uncertainty (Shearer, 2005). Each landscape plan and the expected
changes it addresses is unique for a specific place and for a particular suite of issues and
landscape changes (there is no universal panacea!). Research on landscape planning has
proposed different steps or stages in planning processes. For example Botequilha Leitao
and Ahern (2002) proposed a set of five stages (landscape analysis, diagnosis, prognosis,
44
evaluation and monitoring) while other researchers have suggested from six to eleven
interacting steps (Kato and Ahern, 2008).

Independently of the phases and steps in planning, every landscape plan is subject to the
full spectrum of uncertainty related not only to that specific place but also to neighbouring
areas or even from unexpected events occurring in remote places (Hersperger and Forman,
2003). Dealing with uncertain futures is a facet implicit in planning for sustainability. The
ways in which the future may unfold i.e. “future studies” are the subject of analysis and
may consist of a vast array of approaches of which “scenario” creation and development
are particularly used (Shearer, 2005; Tress and Tress, 2003). Scenarios are not predictions,
instead they are seen as plausible accounts of the future (Carpenter et al., 2006). There are
varying scenario typologies more or less appropriate for different research questions
(Borjeson et al., 2006).

One of the great problems faced in scenario development is the integration of different and
often opposing functions occurring in ecosystems at a multitude of levels (McIntyre and
Hobbs, 1999). It is clear that the use of landscapes for the functions of production e.g. for
cereals or timber, housing, groundwater recharge, recreation and transportation, to cite
only a few examples, creates conflicts that need to be solved through planning (Hanna,
2005; Herrmann and Osinski, 1999). Multi criteria, multi objective approaches are very
important to systematically represent the tradeoffs that need to be made (Kangas and
Kangas, 2005; Malczewski, 2004). Land use allocation procedures in Geographical
Information Systems (GIS) software are also valuable tools in order to account for
different environmental, social and economic options (Kangas and Kangas, 2005; Mendoza
and Prabhu, 2005).

45
Table 1.12.Principles for sustainable landscape planning

Principles Description/methods used Sources

Transdisciplinarity Integrating knowledge from different research (Fry, 2001; Naveh,

fields can help solve and coordinate the 2007; Tress and
conflicting interests when approached as a Tress, 2001; Tress et
common effort by several disciplines. al., 2001)

Uncertainty, knowledge is Adaptability, Multiple lines of converging (Kato and Ahern,

incomplete evidence. Scenario approaches 2008) (Jacobs, 1986)

Multifunctionality Integrating functions such as production and (Brandt et al., 2000;

protection Focus on a number of concurrent Mander et al., 2007;
variables rather than one. Try to optimize Selman, 2002))
instead of maximize. GIS, multicriteria, land
allocation modules

Account for multiple Account for values beyond the marketable (de Groot and Hein,
values of ecosystems 2007)

Equal balance of Linkage between humans and nature Forman (1995 :518)
ecological and Acknowledging social dynamics: culture and
environmental issues traditions are key cohesive forces

Goods and services are Landscapes are more or less sustainable in (Potschin and Haines-
important terms of the outputs of goods and services Young, 2006a;
that are important to people. The tongue Potschin and Haines-
model Young, 2006b)

Spatial arrangement The spatial solution is a pattern of ecosystems (Baskent and

matters! or land uses that will conserve the most Yolasigmaz, 2000;
important attributes such as biodiversity and Blaschke, 2006;
other processes in any landscape. Hersperger, 2006;
Hersperger and
Forman, 2003)

Seek to create new basis Sustainable development strategies should (Matthews and
of attraction and create focus on encouraging “virtuous circles” in Selman, 2006;
virtuous circles landscapes so that the linkages between the Selman and Knight,
socio-economic sphere and environmental 2006)
functions are reinforced.

Engaging the public in the Public preferences are important. There are (Gobster et al., 2007;
design of sustainable differences in preferences amongst Harshaw et al., 2006)
landscapes stakeholders

Address the landscape Environmental and social planning (Hawkins and

scale Acknowledge cross scale interaction Selman, 2002)

Governance for Promotion of partnerships, share of (Jacobs, 1986;

sustainability responsibility between multiple sectors and O'riordan and Stoll-
actors Kleemann, 2002)

Seek technologically Technology is important in developing (Jacobs, 1986)

appropriate solutions “clean” solutions.

Putting this altogether, yet recognising the value of the “no panaceas” rule, (Ostrom, 2007;
Ostrom et al., 2007) there are several concepts, tools, methods and frameworks (Table
1.12 ), often based on transdisciplinarity, which address possible ways to move towards the
46
implementation of more sustainable landscapes in different settings. Potschin and Haines-
Young (2006) propose the “landscapes and sustainability model” to addressing the
sustainable landscapes challenge based on the concept of natural capital. In this model they
argue that landscapes are more or less sustainable in terms of the outputs of goods and
services that are important to people in a specific context. This model was criticized by
Blaschke (2006) because it mistreated the issues related to the spatial arrangement of
natural resources that certainly influence the provision of such goods (Hersperger and
Forman, 2003). The combination of both arguments (goods/services and spatial
configuration) is likely to address two major issues for planning for sustainability,
acknowledging socio-economic and environmental distinctiveness which may be able to
self-reinforce different processes across landscapes. This is further explored in Chapter 4.

Also, work by Selman has suggested that sustainable development strategies should focus
on encouraging „„virtuous circles‟‟ in landscapes (Matthews and Selman, 2006; Selman,
2006) so that the linkages between the socio-economic sphere and environmental functions
are reinforced. The principles behind this argument highlight the need to move from a
vicious circle in which both landscape quality and quality of life for its inhabitants are
deteriorating to a situation in which both are enhanced and self-reinforcing. Figure 1.3
illustrates such a view. It is also argued that there is a need to create “basins of attraction”
(Mathews and Selman 2006:202) admitting that these may vary depending on the socio-
economic and environmental circumstances, highlighting the usefulness of qualitative
approaches for identifying such issues (Selman and Knight, 2006).

Based upon Antrop (2006) and Selman (2006) planning for sustainability can be seen in
two ways. First, planning can be used when the goal is conservation of certain landscape
types or values. Within this logic planning can reinforce the continuation of the existing
practices that maintain and organize these landscapes, independently of whether they are
rural or urban and being of natural, cultural (Box 1.2), or even of ordinary charisma.
Second, planning for sustainability can be used to “guide” the transformations of
landscapes with inherent potential in order to create new basins of attraction able to reach
new virtuous circles (Matthews and Selman, 2006).

Selman (2006) refers to this duality as “old world vs. new world” landscapes challenges in
the following terms:“In old world landscapes, the challenges are essentially those of
finding new and self-sustaining means of retaining landscapes whose qualities are being
undermined by functional obsolescence; in new world, the challenge is often one of
adjusting colonial mindsets to discover new ways (or rediscover old ways) of sustainable
living in fragile and over-exploited terrains” Selman (2006:5)
47
  Outmigration
 Loss of local
entrepreneurship

Landscape quality Vicious Quality of life

circle

 Dissolution of links between

landscape and community
 Loss of distinctiveness

 Vibrant economy and

customs
 Investment in land care

Landscape quality Virtuous Quality of life

circle

 Enhancing personal
well-being
 Land care efforts sustain
population base

Figure 1.3. The vicious and virtuous circles

Source: Selman (2006:172)

The old vs. new landscape challenges apply to the difficulties of dealing with
multifunctionality both at forest and landscape scales (Sections 1.2.2 and 1.2.3). As was
explored in Section 1.1 it has been increasingly acknowledged that some cultural
landscapes are not self reinforcing (Matthews and Selman, 2006). As shown in Table 1.12,
multifunctionality has increasingly been proposed as a principal “hallmark” of landscape,
strengthening its case for being at the heart rather than periphery of integrated spatial
planning (Brandt et al., 2000; Fry, 2001; Mander et al., 2007; Selman, 2002). Also in

48
Section 1.2 it was stressed that it is important to investigate the type of forestry that is
likely to contribute to sustainable landscapes in such a way that man- nature linkages are
self reinforcing.

1.4. Research rationale and aims

The research presented in this thesis examines the issue of how to manage for
multifunctional forests based on forestry management practice in Portugal. In this country
many of mosaics of land use that were viable in the past are not sustainable at present and
rural areas face depopulation trends due to socio-economic changes related to agricultural
and forestry abandonment. Portugal was used as a case study in order to investigate the
ways in which forestry may contribute to more sustainable landscapes. The work started by
using quantitative approaches at the national/regional scales in order to describe the
condition of forests based upon a set of landscape metrics. Subsequently, going down in
scale the research also examined public uses and preferences for forests at the watershed
scale. Going even further down in scale, this time using qualitative approaches, scenarios
for SFM were developed (Figure 1.4).

The broad aim of the research was to inform planning and management strategies for SFM
at different spatial scales. With such an aim, based on a multiscale approach, the study
used concepts and methods from research areas such as landscape ecology and social
sciences, aided by GIS tools. Those tools and methods were differentially used according
to the scale under analysis. Figure 1.4 schematically shows the research framework as well
as the different scales of analysis.

At the national and regional scales, class metrics such as patch size (PS), patch density
(PD) and percentage of landscape (PLAND) for the most common tree species were used
to investigate the ways in which development patterns occurring in the context of Portugal
(increasing urbanisation one hand and land abandonment in the other, see Chapter 2)
affected the composition and configuration of forests. At this stage the tools and methods
used were landscape metrics, statistical analysis and GIS.

49
 Incorporating concepts, tools
and methods from SLP and
landscape governance into SFM
Sustainable landscape planning
(SLP)
SFM

Forest science

Scale Tool/Method

National/regional Landscape metrics

PROBLEM DEFINITION:

How can SLP and landscape River basin Questionnaire survey

governance “guide” SFM through
River basin Photo-questionnaire
landscapes?

Local Focus group

Scenarios for SFM

uestionnaire

“Toolbox” for “guiding” the

implementation of SFM at different
spatial scales ?

Figure 1.4. Research framework

At the catchment scale, the type of forest multifunctionality that is likely to deliver a self-
reinforcing link between people and forests within the urban-rural gradient was studied. In
order to do so, field and questionnaire surveys were undertaken. At the catchment scale
public preferences for forest characteristics were surveyed (by including in the
questionnaire a set of photos showing different forest characteristics)

At the parish scale scenarios for SFM were created and developed for two parishes through
a participatory process which comprised two focus group meetings in each parish and one
final workshop.

50
After answering some specific questions in each chapter (these are summarised at the end
of Chapter 2) the three overall questions presented in Box 1.4 will be used to summarise
the conclusions of this work, namely:

1) How can landscape planning and landscape governance concepts, tools and
methods be used to guide forestry management at different spatial scales? Within this
topic, the usefulness of the set of tools used in tackling the challenge of guiding SFM at
different spatial scales is explored. The usefulness of the toolbox used is described for
three contexts, namely decision making, planning and forestry practitioners.

2) What might be the role(s) of forests in helping to sustain landscapes in Northern

Portugal? One important topic addressed at this stage is the cross-scale coordination
needed to deliver the implementation of SFM at the landscape scale.

3) What are the wider implications of the work developed throughout this thesis?
Under this question the major findings are summarised and implications of this work are
explored.

1.5. Thesis structure

The structure of the thesis is as follows. This introduction raised the topics that need to be
dealt with to implement sustainable forestry management. Chapter 2 describes the
characteristics of the forestry sector in Portugal and sets Portugal within the European
context. At this stage, specific research questions addressed at different spatial scales are
also described. Furthermore, in Chapter 2, the usefulness of different tools to address SFM
in the context of Portugal is further explored. Chapters 3 to 6 are presented in paper format
that I would like to submit to peer reviewed journals. Chapter 3 is already in press in the
Journal of Environmental Management and I intend to submit the other data chapters after
having my viva. Chapter 3 investigates the extent to which forest landscape metrics vary
across different socio-economic settings and identifies planning priorities for rural and
urban areas. In Chapter 4, based on uses and public preferences for forests, the type of
multifunctional forest management likely to reinforce urban-rural partnerships is explored.
Chapter 5 examines multiple functions of forests by contrasting attractiveness and
management public preferences. It further explores the impacts that public preferences may
have on the ecology of forests. In Chapter 6 scenarios for multifunctional forests are
developed and the means to fully implement those are explored. Finally, Chapter 7 gathers
together conclusions and recommendations for future work. Figure 1.5 shows the overall
structure of the thesis.

51
 CHAPTER 1. INTRODUCTION
Sustainable landscape planning &
multifunctional forest management

CHAPTER 2. CASE STUDY

Forestry and land management in Europe:
A case study from Portugal

CHAPTER 3.
Associations between forest
characteristics and socio-economic
development: A case study from Portugal

From CHAPTER 4.
National to
Local scale Integrating public uses and preferences
for forests into multifunctional forest
management plans at catchment scale

CHAPTER 5.
Public opinion regarding attractiveness
and management of forests: preferences
for percentage of forest cover and stand
structures in northern Portugal

CHAPTER 6.
Governing for sustainability:
Implementing multifunctional forest
management in Portuguese rural areas

CHAPTER 7. CONCLUSIONS
The role of multifunctional forests in
sustainable landscapes

Figure 1.5. Thesis structure.

52
 “I like trees because they seem more
resigned to the way they have to live
than other things”
Willa Cather 1873-1947

CHAPTER 2. Forestry and land management in Europe: A

case study from Portugal

53
2.1. Forestry in Europe

The area of forests in Europe has been increasing and protected areas occupy 18 % of
European land (Nabuurs et al., 2001) (Figure 2.1). The physical extent of forest cover in
Europe ranges from 1% in Iceland to well over 50% in Finland or Slovenia (Slee, 2007a).
In the last decade of the 20th century European forests comprised 168.6 million hectares of
the most regularly managed forests and these are amongst the most intensively used forests
in the world (Nabuurs et al., 2001). In Europe, there are only small remnant areas of
relatively undisturbed natural forests such as the forest of Bialowieza in Poland. The
average age of European forests in 1990 was 57 years with a mean growing stock of 142
m3/ha and a net annual increment of 4.6 m3/ha (Nabuurs et al., 2001; Spiecker, 2003).
These mean values hide a huge diversity of forests amongst countries and even within
countries (Niskanen and Lin, 2001; Spiecker, 2003).

Figure 2.1. Forests in Europe

Source: CIFOR (www.cifor.org) [accessed December 2008]

It has been argued by Mather (1992) that forestry in Europe has passed through distinct
historic phases of development: a pre-industrial phase, an industrial phase and a post
industrial-phase. Mather named this trend “the forest transition”. In the pre-industrial phase
54
forestry was principally a provider of local livelihoods supplying a range of timber and non
timber products. Over the 18th and 19th centuries the growing imperial powers and the
industrial revolution created more mono-functional demand for specific types of timber
(the industrial phase). Although this monofunctional industrial style of forestry remains to
some extent in parts of Europe, the most recent post-industrial phase has introduced
demands for more varied styles of forestry with a stronger amenity or post-production
function (Mather, 2004; Mather et al., 1999; Mather and Needle, 1998; Slee, 2007a).

In order to explore the diversity of forests in Europe as well as the diverse ways in which
forests are perceived by policymakers, owners and general public, Slee (2007a: 70) defined
six broad regions, namely: North sea/ North West, Nordic, Mediterranean, Germanic,
Balkan/ Eastern European and Baltic groups. This classification was developed by Slee
through the programme COST ACTION E 30 (Economic integration of urban consumers‟
demand and rural forestry production) .

Table 2.1 summarises the importance given to different functions provided by forests
across the groups defined by Slee (2007a). Within these groups, in different regions,
forests functions are perceived in very different ways by different stakeholders (Elands and
Praestholm, 2008; Elands and Wiersum, 2001) so the importance attributed to each
function in each group is no more than indicative and was based on the description below.

Table 2.1. Forest functions across Europe

Productive Recreational Protective

Timber Non
timber

North Sea/North West + +++ +++ +++

Nordic +++ +++ ++ +++

Mediterranean ++ +++ ++ ++

Germanic ++ +++ +++ +++

Balkan ++ ++ ++ +

Baltic + ++ ++ ++

Comparison of different management approaches with an indication of their respective fulfilment

of different management goals. The goal fulfilment is subjectively scored on a scale from 1 to 3.
“+++”= high goal fulfilment, “++”= medium goal fulfilment and “+”= low goal fulfilment. Source:
Adapted from Slee (2007a)

Countries included in the North Sea/ North West European group (Denmark, Iceland,
Netherlands, UK, northern part of France and Belgium) are generally characterised by high

55
levels of environmental consciousness and a strong interest in forestry and woodland
(Mather et al., 1999; Niskanen and Lin, 2001). The forestry sector in these countries is
characterised by a predominance of non timber production motives such as biodiversity
and recreation (the exception is Ireland which puts some emphasis on production)
(Angelstam et al., 2005; Slee et al., 2004). In these countries, forestry as a productive
activity is mostly grant or subsidy-driven (Nabuurs et al., 2001). The state forest sector in
these countries is often significant and the state is the major provider of forest recreational
activities (Mather, 2004) .

In the Nordic group of counties (excluding Denmark which was included in the previous
group) the prevailing private sector model is of reasonably sized forestry units (averaging
50 ha) (Slee, 2007a). Forests are still seen as an important complementary activity to
farming and strong farm-forest associations support forest productive activity (timber)
(Nabuurs et al., 2001). In parallel to these productive functions there are also strong public
access rights in all forests and wild berries and fungi are widely gathered. Thus, these
countries seek to balance environmental protection and production functions of forests
(Angelstam et al., 2005).

In countries fringing the Mediterranean (Italy, Greece, Spain, Portugal as well as southern
parts of France) management of small scale forests is prevalent (Nabuurs et al., 2001).
Average size of private forest holdings is small and abandonment of forests often occurs
(Pinto-Correia and Breman, 2008). Non timber products are very important in some areas.
There has been growing interest in timber exploitation (Spain, Portugal and Northern Italy)
and industries based on pulp transformation afforest huge areas annually with non native
tree species such as eucalyptus. There are extensive areas of communal lands (in northern
and central Portugal as well as in the Spanish region of Galicia), many of which deliver
important recreational and landscape as well as timber functions (Niskanen and Lin, 2001).

The German speaking countries of central Europe, including Austria, Germany and
Switzerland tend to have a high degree of forest cover (Nabuurs et al., 2001). With the
exception of Switzerland, average holding size is above 10 ha. Both timber production and
non timber forest products are important (Angelstam et al., 2005). There are strong
traditions of multifunctional forest management (Mander et al., 2007; von Haaren and
Warren-Krestzchmar, 2006).

In countries comprising Central and Eastern Europe although restitution of forests to

private holders has happened the State still retains a substantial area of forests which
include important environmentally sensitive areas (Angelstam et al., 2005). Certain

56
activities such as mushroom gathering are important social and cultural activities as well as
delivering subsistence support to poor rural households (Niskanen and Lin, 2001).

Baltic countries were classified as a hybrid between the Nordic and the Central Eastern
models (Angelstam et al., 2005). In these countries there is a desire to create a Nordic
model for holdings, although the holding size is smaller and the farmers have no tradition
of forestry management (Slee, 2007a). The state has retained a substantial proportion of
forests for both wood raw material production and recreational and environmental
functions (Angelstam et al., 2005).

This classification has been presented as indicative of the diversity of forests in Europe.
Within the same group different countries have reported different preferences for
management strategies (Niskanen and Lin, 2001). Even within the same region different
values have been attributed to productive and protective functions of forests in contrasting
places e.g. rural vs urban (Elands et al., 2004; Elands and Praestholm, 2008).

According to Elands and Wiersum (2001), forests in Europe may have five major roles
namely, agri-ruralist, hedonist, utilitarian, community stability and nature conservation
(Table 2.2). Those roles vary from providing income and employment (in the utilitarian
discourse), complementing and diversifying farming systems (agri-ruralist), supporting
socio-economic regeneration of marginalized areas, increasing living conditions in remote
places (community stability) and being preserved as strategic natural areas (nature
conservation).

The different roles of forests will be more or less appropriate for a particular place
depending upon both environmental conditions and the social groups involved. As was
reported by Pinto-Correia and Breman (2008:1) the situation faced in the diverse
conditions of Europe will demand varying types of agriculture and forestry in different
places.

“In some regions, there is a productivist orientation and productive functions have a
dominant economic role, while other [regions] will need to be supported on other
functions to survive economically and socially, or may be best suited to environmental
functions alone. The vocation of the...territories is different, and thus also the functions
they are able to support”.

57
Table 2.2. Rural development discourses

Rural areas Discourses

Agri-ruralist Hedonist Utilitarian Community Nature
sustainability conservation
Conception Farmers as Countryside as Production Remote places Potential nature
stewards of the the garden of areas to be used areas, nature
countryside the city for economic has intrinsic
purposes values
Problem Crises in Deteriorating Under Marginalisation, Uncontrolled
modern aesthetic, development stagnation and incursion of
farming cultural values and retardation decrease in rural areas into
liveability and wilderness
economic
vitality
Future New social Re- Need for Re- creation of Creation of
contract establishment innovative basic social- new controlled
farmers- of these values economic economic balance
society, above all activities structures and between rural
sustainability living areas and
and quality conditions nature areas
Source: Elands and Wiersun (2001)

Exogenous development models in post-war Europe combined subsidizing improvements

to agricultural (and forestry) production to enhance farm incomes, the key principles
focussed on economics of scale and enhancement of production (Okkonen, 2008). The
dynamic force of development was urban growth and the main function of rural areas (of
which forests are an important feature (Niskanen and Lin, 2001)) was primarily production
for rapidly expanding urban economies. This has been considered as causing “dependent”
development; the subsidies, policy decisions and distant agencies, in some cases, meant
erasing the cultural and environmental differences between regions (OECD, 2006). The
problems arising from such exogenous rural development gave an impetus for forms of
endogenous development more focussed on recovering or re- inventing the identity of the
rural territory (Okkonen, 2008).

As was previously discussed in Chapter 1 (Section 1.3), multifunctionality from primary

sector activities such as agriculture and forestry has been increasingly valued in post-
industrial societies (Pinto-Correia and Breman, 2008). There are however different
circumstances across Europe and this diversity has been addressed by different research
projects such as EU RURALIS and RUFUS- rural future networks. The diversity of
circumstances across Europe also calls for different types of functions occurring in
contrasting settings. Pinto-Correia and Breman (2008) propose a methodological approach
to identify different vocational areas in which different types functions of agriculture and
forestry are likely to occur.

58
In a much differentiated Europe, in contrast to agriculture, forests do not have a European
policy framework. Instead, forest policies are generally included in the Common
Agriculture Policy (CAP). However, other measures put into place by the EU such as
carbon markets and biodiversity are of great importance for the forestry sector. The EU
identifies a clear role for forestry in combating climate change and as major contributor to
the protection and maintenance of biodiversity (Dwyer, 2007). Currently, the CAP is under
transformation towards a Common Agricultural and Rural Policy for Europe (CARPE),
within this protective forest functions have been receiving much attention. In forestry and
woodland management emerging market trends for forestry biomass and carbon
sequestration appear potentially beneficial, particularly for Europe‟s rural environment
(Dwyer, 2007). Although this is monitored by the Ministerial Conference for the Protection
of Forests in Europe (MCPFE) the responsibility for forestry policies lie with the
individual Member States.

Different European countries have distinct planning systems but there is an overall trend
not to include activities such as forestry into formal planning mechanisms (Niskanen and
Lin, 2001; Selman, 1997). Instead, there are very detailed sectoral forestry plans that have
not been especially successful at integrating forestry within the wider landscape context.
Moreover, some of these forestry plans are not mandatory (CBD, 2008). Above all, it has
been reported that multifunctional management plans are missing (Selman, 2002).
Germany is one of the EU countries which has been focusing on sustainable landscape
planning and its land planning system is an example for most of the remaining EU
members (von Haaren and Ott, 2008). In the formal planning system in Germany, the State
as well as the regional and local Governments are obliged to produce a landscape plan (von
Haaren and Ott, 2008). However, in most States the landscape plan is not, in principle,
legally binding only becoming so when integrated into town and country planning (von
Haaren and Warren-Krestzchmar, 2006). Town and County planning in the UK is based on
the production of development plans which zone different categories of use. Some land
uses are regulated by statute (including construction, mineral extraction) whereas other
land use categories (e.g. forestry) are not directly regulated (Selman, 1997, 2002).

The problems arising for framing strategies for SFM in Europe are therefore twofold 1)
there is a need to integrate and coordinate forestry policies across Member States in order
to explore the emergent markets of goods and services from forests and ii) each Member
State attempts to develop forestry policies, often without having forests integrated in the
formal planning system. Furthermore, there are enormous differences in forest cover across

59
Member States in Europe, with even the amount and condition of forests being highly
variable within a single country (Elands and Wiersum, 2001; Niskanen and Lin, 2001).

It follows that across Europe post productivist conditions call (now more than ever) for the
integration of environmental considerations into policy. In forestry and woodland
management this means a shift from timber based production to “production” of other
environmental goods (Sayer and Maginnis, 2005). Although the EU identifies a clear role
for forestry in combating climate change as well as being a major contributor to the
protection and maintenance of biodiversity (Dwyer, 2007), the EU puts faith on isolated
measures hoping that the individual Member States will manage forests sustainably as is
the goal of the Ministerial Conference on the Protection of Forests in Europe (MCPFE).

2.2. A case study in Portugal

As previously stated Portugal will be used as a case study throughout this thesis. In the last
few decades major socio-economic changes have been occurring in this country. Forests
are an important landscape feature whose management has faced a number of challenges.
There are major threats to sustainability arising from wildfires, an increase in area of non
native tree species such as eucalyptus and a steady increase in abandonment of the inland
rural areas. Meanwhile major city regions located near to the sea coast such as Lisbon and
Porto are becoming increasingly urbanised (Figure 2.2).

Given the importance and diversity of forests in Portugal, as well as the range of socio-
economic settings, I felt that the country provided an appropriate example in which to
investigate issues of sustainable development with regard to forestry.

2.2.1. Brief description of the country

The area of the Portuguese mainland outlines the shape of a “rectangle” located in the
Iberian Peninsula, in the far southwest of Europe and covers for 88,889 km2 (218 km in
breadth, 561 km in length). It has 832 km of Atlantic coastline and 1,215 km of border
with Spain (INAG, 2006; INE, 2008). Its physical environment can be generally described
as mountainous north of the Tagus River with plains in south (Figure 2.2).

60
 Figure 2.2. Portugal: Location and topography

Source: Maps of the world (www.united-states-map.org) [accessed 12 March 2009]

Portugal has two climatic regions: the region under influence of the Atlantic sea, and the
inland region which has a Mediterranean climate. The mean annual temperature varies
between 7ºC in the hills of the interior central region to 18ºC on the South coast. The
North part of Portugal has two major regions. In the northwest part the “green Minho” is
located (study area of Chapters 4, 5 and 6), in the northeast the Tras-os-Montes region is
situated. South of Minho and Tras-os-Montes are the central mountains of Serra da Estrela
of which the highest peak is 1,990 metres. Located in the southern part of Portugal are the
plains of Alentejo dominated by “montado” and the well known tourist region of Algarve.
Montado, which corresponds to the Spanish dehesa, is a characteristic agroforestry system
based on livestock (Pereira and Fonseca, 2003; Pinto-Correia, 2000).

The rainfall across the Portuguese mainland greatly varies in space and in time. Mean
annual rainfall values greater than 1500 mm occur on the coast in the North of the country
and in the hills of interior regions in the North and Centre. The mean annual rainfall is
about 920 mm in the Portuguese mainland having a non-uniform spatial distribution more
concentrated on the Northwest (Minho) hydrographical region where the rainfall level is
61
practically double the average for the remaining country (Moreira et al., 2001b). Average
rainfall distribution in the year is markedly non-uniform: about 75% to 80% of the mean
annual rainfall is concentrated in the six wettest months, from October to March. Figure
2.3 shows four maps in which elevation (upper left), mean annual temperature (upper
right), soil types (bottom left) and major river basins in the Portuguese mainland are
represented.

Figure 2.3. Portuguese mainland characterisation

62
Demographic indicators for 2007 reveal that the main recent demographic trends in
Portugal have remained unchanged: slower population growth and demographic ageing
(INE, 2008). The natural growth rate, which has been showing a decreasing trend for
decades, displayed in 2007, for the first time in recent Portuguese demography, a negative
value (-0.01%). Since the beginning of the 20th century on only one other occasion, in
1918, was a negative value registered, due to the 1918 pandemic flu that affected the
country (INE, 2008).

Portugal‟s territorial organization consists of 18 Districts in the mainland and two

autonomous regions; the Azores and Madeira islands. Inside this administrative and
territorial organization Portugal is a small but rather diversified country in terms of
landscape (Firmino, 1999). Portugal is a mosaic of landscapes, partly because of its
different climatic conditions and morphologic genesis but also because of the impact of
human activity on land use. In Portugal, as happens elsewhere in Europe, a polarisation of
human and economic resources along the sea coast is found (Antrop, 2004). The poorer
Portuguese territory is located in the interior of the country along the border with Spain.
Close to the border areas of Spain the “deep countryside” can be still found where natural
landscapes and culture survive relatively intact (Firmino 1999).

There are studies that report that during the last two decades of the 20th century Portugal
experienced a period of increasing material consumption which was associated with
transitional economies (Canas et al., 2003; Niza and Ferrao, 2006). Studies such as that by
Niza and Ferrao (2006) demonstrated that the development model that led the Portuguese
economy to a transitional state corresponded with an increase in material consumption
related to strengthening the critical infrastructures (e.g. roads) that were required to
improve transportation and housing. In particular, the study shows that the highest growth
rate occurred in the construction sector. This “consumptive period” was followed by a
period in which the Portuguese economy became more oriented to service sectors and,
predictably, to less material intensity. Despite efforts made to increase the average
schooling years, there are still issues related to educational quality that need to translate
into an increase in labour force skills (Pina and Aubyn, 2005). During the “consumptive
growth” period rapid land use change and environmental problems arose.

2.2.2. Forestry and land use planning in Portugal

The vast majority of the landscapes in the country before the 1940s could be characterized
as follows: the valleys and shallow slopes were used as agricultural land, with extremely
small mixed farm holdings (usually less than 2 ha) with pastures and a wide range of crops

63
such as corn, cereals and potatoes. Surrounding the valleys, in the slopes between the
farmed land and the mountains, there was a region where privately owned shrubland areas
with some scattered trees (called boucas) as well as communal small plots (called
maninhos) were used as a source of timber and organic matter to fertilize the lowland soils.
At higher altitudes, large extents of communal lands (called baldios) were used as pasture
for cattle, sheep and goats. In the baldios small to medium sized forested patches of pines
and oak also occurred (Moreira et al., 2001b). The photo in Figure 2.4 shows a typical
Portuguese landscape that still prevails in the deep countryside of the Northern Region of
Portugal.

Figure 2.4. A typical rural landscape in Minho region of Portugal

Several changes in socio-economic and political conditions in the second half of the 20th
century caused abandonment of farming activities and emigration flows which were most
pronounced during the 1960s (Moreira et al., 2001b). Parallel to this decline in population
and agricultural activities, several policy measures (Table 2.3) promoted afforestation on
communal lands.

Denoting the long lasting importance of the forestry sector in Portugal the Forestry
Institute was created in 1886 (Table 2.3). The Forestry Regime legislation (1901) is one of
the most important in Portuguese forestry history. Inspired by French legislation with the
same name this law is defined as “implementation of legislation mechanisms to assure
whether the creation, exploration and conservation of forests according the national

64
economy whether to defend the public utility as safeguarding the erosion, the water system
and other environmental aspects”.

In addition to the Forestry Regime the “Plano de Povoamento Florestal” aimed at

afforestating the communal lands (baldios) as well as shrubland areas in the slopes and
mountains and the “Fundo de Fomento Florestal” aimed at promoting afforestation of
private land. Other political measures included the “Projecto Florestal Portugues” and the
“Programa de Accao Florestal”. All these policy instruments promoted afforestation
mostly based on coniferous trees, particularly maritime pine (Pinus pinaster) (Moreira et
al., 2001b). Table 2.3 shows the different change in strategies that have been occurring in
forestry organisation.

Landscape planning approaches in Portugal started to be used during the 1960s although
only by a small minority of professionals involved in urban and regional planning. In the
following decade (1970s) environmental degradation became a significant problem as a
result of plans for development which emphasised paper production, textile and chemical
industries as well as other heavy industries.

In addition to these factors, increasing tourism pressures, particularly along the coast, led
to rapid land use changes. In order to steer away from these unfavourable trends of land
use change, landscape planning shifted from a hobby of a minority of professionals to a
valuable tool to support environmental policies. As a result, in the 1970s, natural reserves,
the only national park (Peneda-Geres National Park) and other sites and protected
landscapes were classified.

In the years 1982 and 1983, the Portuguese government created four mandatory landscape
plans namely, National Agriculture Reserve, the Town Master Plan, the National
Ecological Reserve and the Land Use Regional Plan (Table 2.4). The former two were
created in 1982 and their goal was to protect 12% of prime land suitable for agriculture
(National Agriculture Reserve) and to define the principles and rules of land use as well as
the integration of the plans with other planning levels (Town Master Plan). Created in
1983, the National Ecological Reserve and the Land Use Regional plans aimed at
safeguarding fragile ecosystems and heritage landscapes threatened by development as
well as defining land-use norms for regions economically and ecologically homogeneous.

65
Table 2.3. Forestry laws and institutional organizations in Portugal

year Law/Institution Function

1886 Creation of Forestry The forestry services were included in Direccao

services Geral de Agricultura dec 25/11/1885

1901 Forestry Regime Manuel Francisco Vargas created the special fund
for forestry services

1936 Creation of JCI Creation of Junta de Colonizacao Interna in order

to increase productivity both in agriculture and
Law 27 207 forestry sector
11/11/1936

Law 1971 15 June Afforestation plan Afforestation of 420 000 ha

1938

1976 Portuguese constitution Portuguese constitutions which refers to forests as

an national interest

19/01/1976 Law 39and 40 of 1976 Law regulating the communal land management
systems

1965 Fundo de Fomento Promoting afforestation of private land

Florestal

1980s PFP and PAF Projecto Florestal Portugues Programa de Accao

Florestal

1995 Law nº 134/95 Approval of the Peneda-Geres National Park

management plan

17/08/1996 Law n.º 33/96 Forestry law

15/03/1998 Law n.º 33/98. Creation of an agency to manage public and

communal lands Empresa Pública Florestal
ENGEF.

08/04/1999 Law n.º 27/99 of 18th of Sustainable plan for the development of
March. Portuguese forestry

09/06/1999 Law n.º 204/99 and205/99 Regulation for regional forestry management
plans (PROF) and local plans (PDF)

28/07/1999 Law n.º 14307/99. Extinguishement of ENGEF

14/04/2001 Law n.º 7781/2001 Creation of a COFLORGEST- Commission for

the management of public and communal forests

2003 Creation of DGRF Extinguishment of DGF and creation of DGRF

2008 Extinction of DGRF Creation of Autoridade Florestal Nacional (AFN)

Creation of AFN

From 1986 up to 1999 the “development” across the Portuguese territory was financially
supported by the EU Regional Development Plan (Portugal Joined the EU in 1986) namely
through the European Regional Development Fund (ERDF), the European Agriculture
Guidance and Guarantee Fund (EAGGF) and the European Social Fund (ESF). At the
66
same time, the Law of Environmental Base created in 1987, helped to enforce awareness of
environmental issues. In the same way, the fourth environmental program (1987-1992)
tackled environmental issues such as prevention of pollution and the improvement of
natural resources management. This program was, however, silent regarding landscape
planning as a tool to support environmental policy. Despite such efforts on environmental
issues economic development funds such as ERDF and EAGGF had major impacts on the
Portuguese landscapes (Andresen and Castelbranco, 1993).

In Portugal the integration of forestry with other land use planning systems is weak
(Fidelis and Sumares, 2008). The formal land use planning system is the responsibility of
the Environment, Territorial Planning and Regional Development Ministry (Ministerio do
Ambiente, Ordenamaneto do Territorio e Desenvolvimento Regional MAOTDR). Forestry
planning responsabilities are included in the Agriculture, Rural Development and Fisheries
Ministry (Ministerio da Agricultura Desenvolvimento Rural e das Pescas MADRP). The
two Ministries, in theory, should cooperate but in practice there are few linkages between
them. In addition to this, forestry in Portugal has a top-down type strategy. The Forestry
Policy Act (1996), provides for the national strategy for forests in Portugal, together with a
Plan for the Sustainable Development of the Portuguese Forest PSDPF (1999). Also at
national level there is a plan for protecting forests against fire (PNDFCI). The relevant
instrument in the Portuguese Constitution states that “the State will promote forestry
policies according to ecologic and social circumstances” (Portuguese constitution, 93rd
article, number two). At national level there is also a funding scheme created from
revenues from petrol consumption (Fundo Florestal Permanente) which provides financial
support for forestry related investments (DGRF, 2007b).

At regional level, in addition to PROFs (forestry sector plans), there are the PROTs
(Regional Plans that regulate all land uses), and the PEOT (created exclusively for
regulation of land use allocation in protected areas). All three regional plans are only
mandatory on public land which represents approximately 2 % of all forest land. PMOTS
are the mandatory plans at the local level for private and communal property and include
the Municipal Director Plan (PDM). This regulates all land uses, the urbanisation plan
(PU), and other specific plans (PP). The Plano Director Municipal (PDM) is the landscape
plan which incorporates the municipal plan for defence of forests against fire (PMDFCI).
Table 2.4 shows the plans and the regional scale addressed. These are the official plans
that, in 2008, ruled the forestry sector. There have been, however, continuous changes in
the “strategy” of forestry polices in Portugal and there is a need to better integrate forests
within the formal planning system.
67
Table 2.4. Current arrangements for planning and funding forest management in Portugal

Scope Formal Planning MAOTDR Forestry planning MADRP

National Portuguese constitution
National Plan for “land use planning”
PNOT
National Agriculture reserve (RAN)
and National Ecological Reserve
(REN)
Forestry Policy Act (1996)
Portuguese strategy for forests
(EFN)
Plan for the Sustainable
Development of the Portuguese
Forest PSDPF (1999)
National plan for defence of
forests against fire
PNDFCI

Regional Regional plan for ordenamento do
territorio PROT
Special plan for ordenamento do
territorio (protected areas) PEOT
Municipal Municiplal plan ordenamento do
territorio PMOT
Municipal directive plan PDM
Urbanisation Plan (PU)
Local
Regional Plan for forests PROF
Municipal plan for defence of
forests against fire PMDFCI
Management plan for private
forests PGF
Management plan for communal
forests PUBs
Management plan for private and
communal forests ZIFs

In 1993, Andresen and Castelbranco questioned the likelihood of a post-modern landscape

being achievable in the short term “the gap between landscape planning, environmental
policy and economic development raised serious questions about the likelihood of a post-
modern landscape being achieved in a near future” Andressen and Castelbranco
(1993:187). At the time their work was published (1993) their concerns had to do with the
pressure of environmental legislation that was still fragile and relatively untested. Sixteen
years on (in 2009) the Town Plan as well as national agriculture and ecological reserves
(although with several changes in legislation) have had time enough to establish roots,
however, land use changes driven by “economic development pressures” are still an
overwhelmingly obvious reality in Portugal (Fidelis and Sumares, 2008). One of the
problems that was raised by Andresen and Castelbranco in 1993 was the fact that
municipalities receive their funds based on criteria such as number of inhabitants or area
occupied by infrastructure such as roads and airports (which is understandable because
these require major financial expenses). However, because these criteria do not account for
“naturalness” or “uniqueness of the area” municipalities do not have incentives (except
from tourism revenues that anyway are not direct) to preserve either agriculture nor

68
heritage landscapes. Fidelis and Sumares (2008: 299) described the situation in Portugal at
the present as follows:

“Natural areas and associated biodiversity continue to be degraded by the negative

impacts of urbanisation, which translate into altering natural drainage patterns and
natural rainfall-runoff-storage relationships, fragmenting habitats and adding non-point
pollution and pollutants to runoff and streams, increasing surface temperatures that cause
decline in habitat quality and biodiversity and limiting the public’s ability to enjoy many of
the benefits these area provide...”

As is evident in Table 2.3 on several occasions, land management and forestry regulations
have been changed yet, in spite of integration of planning throughout the whole landscape,
emphasis has been put on sectoral approaches. There are agriculture, urbanization and
forestry specific programs created by autonomous (public) organisations in which the links
with regional planning are loose (Fidelis and Sumares, 2008). The creation of autonomous
organisations (such as agencies) in government has been a trend in many OECD countries
influenced by „new public management‟ ideas on methods of how to organise the public
sector (Araújo, 2001). The arguments for this convergence of administrative reforms are
that new organisational forms will, in theory, „work better and cost less” (Araújo, 2001;
Silveira, 2000). Araújo (2001:3) examined the influence of new public management ideas
on recent changes in Portuguese central administration and concluded that public
management “is not restricted to effectiveness and efficiency reasons but, above all, by the
values of the public domain and politicians”. Forestry is an important sector in the
Portuguese economy and the public has been feeling the consequences of less successful
management. Forest fires, for example, are a matter of concern for the whole society but
still there are not yet effective solutions for such an enormous problem. The following
section further describes the Portuguese forestry sector.

2.2.3. Moving towards SFM in Portugal

The total forest area in Portugal is estimated at 3.24 million hectares (Rego 2006). This is a
similar value to the National Forest Inventory carried out in 1995-1998 (3.20 million
hectares). According to official Portuguese sources the total forest area corresponds to
around 38% of the Portuguese land cover. Estimates in international databases such as
OECD slightly increase the area of Portuguese forest cover to around 40 %. The
methodologies and technology to produce these estimates are different, thus, the values
should be looked at from a qualitative perspective. As can be seen in Figure 2.5 Portugal is
one of the OECD countries in which forest cover is higher.

69
 80.00%
70.00%
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%
0.00%

Korea, Republic of …

Greece

Luxembourg
Mexico
Austria

Netherlands

Ireland
Turkey
United Kingdom
Japan

Belgium
Sweden

Australia
Italy

France

Iceland
Switzerland

Denmark
Slovakia
Finland

Portugal

Germany

Poland

Spain

Canada

Hungary
Norway
New Zealand
Czech Republic

United States
Figure 2.5. Forest cover in OECD countries

Source: http://dataranking.com/table.cgi?LG=e&TP=ee01-4&RG=1 [accessed January

2009]
According to the last two forestry inventories (1995/98 and 2004/2006) maritime pine
(Pinus pinaster Ait.), cork oak (Quercus suber L.), eucalyptus (mainly Eucalyptus
globulus Labill.), holm oak (Quercus ilex L.) and stone pine (Pinus pinea L.) together
make up 91% of Portuguese forest cover (DGRF, 2007a). Maritime pine is the main
conifer (29%) and is spread across the country but predominates in the centre and north.
The area occupied by eucalyptus represents 23% of the forestry area and has been
increasing steadily (DGRF, 2007a). Forest structure and composition varies across the
country (Figure 2.6). In the South, the previously described montado (broadleaved trees
such as cork oak and holm oak) are a predominant landscape feature. In the North and
Centre maritime pine planted in the 40s (several afforestation plans in Table 2.3) as well as
common oak (Quercus robur) are the most representative, normally in mixed stands.

Portuguese forest land is mainly in private ownership (80%), spread over more than
400,000 holdings of which the vast majority ( 93% of the holdings) are smaller than 10 ha,
(DGRF, 2007a, b). Pulp industries own 6% of the forest area, local communities 12%
(baldios) whilst the forest public area is only 2% (DGRF, 2007a, b).

Fire is a major issue for forestry management in Portugal (Moreira et al., 2001a). It is
known that both fire frequency and area burnt has increased since the mid 20th century,
along with rural exodus and decreasing intensity of agricultural uses (Fernandes, 2001;
Moreira et al., 2001b). Contrary to other southern European countries, burnt area in
Portugal has increased in the last decades (Catry et al., 2005). A study by Catry et al.

70
(2005) for the period 2001-2005 showed that many fire ignitions are located in the most
populated parts of North and Central littoral region. Although municipalities with more
than 100 inhabitants only represent 21% of the territory they accounted for more than 70 %
of fire ignitions and about 14% of total burnt area. The extent of area burnt is higher in less
populated areas, mainly in inaccessible mountainous regions (Moreira et al., 2001b). The
majority of fire ignitions (98%) occur within 2 Km of urban areas (Catry et al., 2005).
These results emphasise the crucial role of human distribution and activity in the spatial
distribution of wildfire ignitions (Moreira et al., 2001a). They also highlight the need to
integrate forest planning, management and governance within a landscape context because
one of the major threats to forest sustainability arises from its proximity to urban land use
(Clemente et al., 2005; Fernandes, 2001).

Figure 2.6. Stand composition (broadleaves, coniferous, mixed stands) in Portugal

71
In summary, the problems concerning forestry in Portugal arise from:

1. The “consumptive” development process, which at the national scale affects forest
land,
2. A formal planning system which barely includes forestry issues,
3. The small size of most forest property holdings making economic exploitation
difficult,
4. Depopulation trends in rural areas with coastal regions becoming increasingly
urbanised,
5. The wild fires which almost every year destroy vast areas of forestry and
unfortunately cause fatalities.

In order to tackle these issues there is a need to implement SFM in Portugal. At the
present, the public as well as governmental bodies are increasingly aware of the need to
move towards SFM (MAOTDR, 2007). Much has been done (e.g. public education for
tackling forest fires, training courses to enhance competitiveness in the forestry sector) but
the lack of effective planning as well as the continually changing governance systems in
the forestry sector has its effects on the ground (Fidelis and Sumares, 2008; Santos et al.,
2001).

This research examines possible ways to move towards the implementation of SFM and
hence aims at exploring the ways in which forestry may contribute to more sustainable
landscapes in Portugal. As was described in the introduction, attaining sustainable
landscapes requires a transdisciplinary team (integrating the knowledge of different
research fields as well as engagement of the public). This thesis was the work of a single
researcher and therefore cannot be transdisciplinary in the strict sense. Instead, a
multidisciplinary approach was used by putting together concepts, tools and methods from
different research fields within a framework, and analysing specific questions at different
spatial scales in a way that the higher levels were informative for the analysis at scales
below (Herrmann and Osinski, 1999).

2.2.4. Research questions

The approach used throughout this PhD addressed SFM in the context of Portugal at
different spatial scales. As was explained in the Chapter 1 varying tools were employed at
different spatial scales. At the national scale because there were generalised “pressures” in
the Portuguese development patterns (as referred by Niza and Ferrao 2006) this work used
landscape metrics in order to investigate if different socio-economic areas had contrasting
forests characteristics (Chapter 3). Subsequently, the research surveyed uses and public
72
preferences at the level of watersheds which are said to be appropriate to explore landscape
scale issues. Finally, at the parish scale ways to move towards more sustainable forest
management were investigated by developing scenarios through a participatory process.
Because public opinion is of great importance in forging strategies for the forestry types of
the future, the research included public involvement throughout (3 out 4 data chapters
involved different forest stakeholders as well as the general public). In order to do so the
use of visual tools (e.g. photography) was widely used due to its ability to effectively help
in engaging stakeholders for sustainable forestry management (Lewis and Sheppard, 2006;
Meitner et al., 2005; Sheppard, 2005; Sheppard and Harshaw, 2000, 2001)

As indicated in Chapter 1, at broader scales there are several studies focussing on

landscape metrics of forests but so far, to my knowledge, none have investigated the
variation in landscape metrics in diverse socio-economic settings (Botequilha Leitao and
Ahern, 2002). This provided an avenue for conducting an innovative piece of research.

At river basin scale, thus far, the sustainable landscapes framework (despite highlighting
the need for creating synergies across regions) does not show obvious ways in which such
an issue can be tackled (Antrop, 2006). At the watershed scale there are several studies
addressing public preferences for the natural environment but only few contrasted scenic
beauty preferences with preferences for management strategy (Sheppard and Harshaw,
2000, 2001).

At local scales two key issues are to define designs for more sustainable forests in which
stakeholders views are represented and explore multiple functions of forests (Elands and
Praestholm, 2008; Elands and Wiersum, 2001). Moreover, if designs and planning
principles are agreed by stakeholders, how can the knowledge developed from landscape
governance help to implement “successful” forestry planning on the ground. These
questions are explored in the research undertaken in the following chapters. Figure 2.7
shows the different scales of analysis as well as the specific research questions addressed.

73
 Chapter 3
National /regional
1. Are there relationships between forest
condition and measures of socio-economic
development?
2. Is there a systematic variation in forest
metrics according to the stage of socio-
economic development of the municipalities?
3. How can landscape metrics from forests
inform strategies to enhance sustainability?

Chapter 4
1. What role(s) should forest serve in the study
River basin
area?
2. Is the forest role(s) identical across the area
(e.g. in urban and rural areas)?
3. How can forests contribute to the sustainable
development of the whole region?

Local (parish)
Figure 2.7. Research questions
Chapter 5
1. Do public preferences for forests vary
according to whether attractiveness or
management objectives are considered?
2. Are there differences in aesthetic and
management preferences across different user
groups?
3. Is there any indication that human
preferences may threaten the ecology of
forests?
4. Do public preferences vary according to
whether verbal or visual approaches are used?
Chapter 6
1.What scenario storylines are considered
viable?
2.What type(s) of planning approaches and
governance systems are needed to implement
multifunctional forests?
3. How can these two storylines be
implemented through a pilot- scheme?
Figure: research questions
74
Chapter 3. Associations between forest characteristics and
socio-economic development: A case study from Portugal

The content of this Chapter is published in the Journal of Environmental Management. The
publication is presented in Appendix 1.

75
Abstract

The integration of socio-economic and environmental objectives is a major challenge in

developing strategies for sustainable landscapes. We investigated associations between
socio-economic variables, landscape metrics and measures of forest condition in the
context of Portugal. The main goals of the study were to 1) investigate relationships
between forest conditions and measures of socio-economic development at national and
regional scales, 2) test the hypothesis that a systematic variation in forest landscape metrics
occurs according to the stage of socio-economic development and, 3) assess the extent to
which landscape metrics can inform strategies to enhance forest sustainability. A ranking
approach and statistical techniques such as Principal Component Analysis were used to
achieve these objectives. Relationships between socio-economic characteristics, landscape
metrics and measures of forest condition were only significant in the regional analysis of
municipalities in Northern Portugal. Landscape metrics for different tree species displayed
significant variations across socio-economic groups of municipalities and these differences
were consistent with changes in characteristics suggested by the forest transition model.
The use of metrics also helped inform place-specific strategies to improve forest
management, though it was also apparent that further work was required to better
incorporate differences in forest functions into sustainability planning.

Keywords: forest sustainability, forest transition, landscape metrics, landscape planning

76
3.1. Introduction

Implementing sustainable development at spatial scales from the international to the local
is nowadays a key goal for many researchers, planners, governments and non-
governmental organisations. Since the Brundtland report defined sustainable development
as “development that meets the needs of the present without compromising the ability of
future generations to meet their own needs” (WCED, 1987:8) there has been an extensive
discussion in the literature about the integration of socio-economic and environmental
issues in order to attain sustainability (Antrop, 2006; Lele, 1991; WCED, 1987)

Recent work has suggested that sustainable development strategies should focus on
encouraging “virtuous circles” in landscapes (Matthews and Selman, 2006) so that the
linkages between the socio-economic sphere and environmental functions are reinforced.
However the form of these “virtuous circles” varies between rural and urban areas. Selman
(2006) cites Antrop (2004) as arguing that landscapes in Europe “can broadly be
categorized as urban centre, urban fringe, rural of urban and deep rural, and that these
display characteristic structures, functions and conflicts” (Selman, 2006:146). In an urban
area where land use is becoming more intensive the policy emphasis is likely to be on
“guiding” processes of change, while in rural regions experiencing depopulation there is a
need to focus on economic and social regeneration (Antrop, 2006).

Landscape ecology offers theories and methods that can contribute to the formulation of
sustainability strategies through a better understanding of processes and functions in
different environmental settings (Potschin and Haines-Young, 2006b; Wu, 2006). A key
tool in landscape ecology is the use of metrics that describe the spatial structure of a
landscape in terms of both composition and configuration (McGarigal et al., 2002). A
number of researchers (e.g. Botequilha Leitão and Ahern 2002) have discussed possible
relationships between landscape metrics and sustainability (see summary in Table 3.1), but
it is generally recognised that there needs to be more empirical assessment of such
associations, particularly at different spatial scales.

This paper aims to contribute to the ongoing landscapes and sustainability debate (Antrop,
2006; Blaschke, 2006; Potschin and Haines-Young, 2006b; Wu, 2006) through an
examination of relationships between indicators of socio-economic development,
landscape metrics and measures of forest condition in Portugal. There have been major
socio-economic changes in this country since it joined the EU in 1986 and forests are an
important landscape feature whose management has faced a number of challenges
(Firmino, 1999; Pinto-Correia, 2000).

77
Table 3.1. Associations between landscape metrics and sustainability

Metric type Metric Relation with sustainability

Area/density/edge Percentage of “If one class dominates completely the landscape

metrics landscape (PLAND) then it will provide little support for multi-habitat
species” (Botequilha Leitao and Ahern, 2002) p75
“At its lowest limit, there is only one land use type
and landscape lacks diversity” “The arrangement of
coarse/fine grained areas within the landscape is
doubtless key factor to achieve a sustainable
environment” (Forman, 1995) p 489.

Number of patches “If mean patch size is small and number of patches
(NP), Patch size (PS) is high it can indicate a fragmented landscape”
and Patch density (PD), (Botequilha Leitao and Ahern, 2002) p75
Patch richness (PR)
“To increase sustainability the obvious solution to
shortcomings of both coarse/fine grained
landscapes is to vary in grain size” (Forman, 1995)
p491.

Diversity Patch richness(PR) “The heterogeneity provided by patches and

corridors in an area plays a key role in
sustainability” (Forman, 1995) p 488.

Shape metrics Perimeter area ratio “Heterogeneity per se appears useful to planning a
distribution (PARA), sustainable environment, but more important is the
actual arrangement of patches and corridors”
“Geometry patterns are indicators of human
disturbance (roads, urban areas)” (Forman, 1995)
p489.

Isolation/proximity Nearest neighbour “greenways offer a promising planning strategy to

distance (MNND) address the challenge of making landscape
Proximity (PROXIM) planning sustainable” (Ahern, 1995) p152. “The
spread of disturbances such as diseases and fire are
greater when MNND is low and when PROXIM
values are high” (Botequilha Leitao and Ahern,
2002) p75. Consensus is emerging: some form of
ecological infrastructure is necessary to achieve a
sustainable landscape condition (Rescia et al.,
2006).

Contagion/interspeciation Contagion, dominance, Indicators for landscape stability and resilience at

and Isolation/proximity Fractal dimension, water catchments level were developed in order to
metrics Lacunarity, Diffusion describe/represent condition and trends of change
rates, Percolation in water catchments with the goal to manage
towards more sustainable condition (Aspinal and
Pearson, 2000).

Largest Patch index Indicators for change in landscape structure caused

(LPI) , Fractal by urbanization provided information about
dimension index specific aspects of landscape structure and thus
(FRAC) Euclidean were helpful to “guide” process of urbanization
nearest neighbour towards sustainability (DiBari, 2007; Ji et al., 2006)
(ENN)

78
At present, some 38% of land is occupied by forests, but there are major threats to
sustainability arising from wild fires and an increase in the area of non-native tree species
such as eucalyptus (DGF, 1996). Rudel et al. (2005) suggest that economic factors linked
to labour scarcity have been a key driver of the forest transition in Portugal and at present
the stage reached (pre-industrial, industrial or post-industrial, see Mather, 1992) appears to
vary across the country (particularly on a transect from the coast to the inland mountains)
(Rudel et al., 2005). Comparing the 1995 and 2005 forest inventories indicates that many
coastal areas had changes in forest area in the range -10 to +10%, whereas adjoining the
Spanish border the values were typically -30 to -50% (DGRF, 2007a). There are still rural
areas with common lands where forests provide a variety of products typical of the pre-
industrial stage, as well as districts where timber production dominates and other urban
regions or national parks where the service functions characteristic of post-industrial
forestry are apparent. This diversity in turn raises questions as to how criteria for
sustainable forest management should be defined (Shifley, 2006). Many different
approaches to this issue exist including the application of indicators for Sustainable Forest
Management (Kangas and Kangas, 2005; Lopes-Ridaura et al., 2005; Munda, 2005) the
adoption of practices such as continuous cover forestry and “back to nature” management
strategies (Gamborg and Larsen, 2003), the use of certification tools, and the valuation of
goods and services provided by the forests (Mangold, 1995; Sheppard, 2005). However,
the use of such techniques needs to be sensitive to socio-economic circumstances so that
the “virtuous circles” underpinning sustainable development are reinforced.

Given the importance and diversity of forests in Portugal, as well as the range of socio-
economic settings, we felt that the country provided one appropriate example in which to
investigate issues of sustainable development with regard to forestry. In developing our
analysis we sought to consider both national and regional scales and to combine variables
from several different research fields (e.g. landscape metrics and socio-economic
characterisation) to bridge the social, economic, and environmental dimensions of
sustainability. The need for this type of integrative effort (interdisciplinary or
transdisciplinary) has been increasingly recognised in studies of environmental and
landscape change (Tress and Tress, 2001) and we see our research as an example of this
wider approach.

Three specific goals arose from these considerations, namely to 1) investigate relationships
between forest conditions and measures of socio-economic development at national and
regional scales, 2) test the hypothesis that a systematic variation in forest landscape metrics

79
occurs according to the stage of socio-economic development and, 3) assess the extent to
which landscape metrics can inform strategies to enhance forest sustainability.

3.2. Data and Methods

3.2.1 Data sources

We used data on forest and socio-economic characteristics at national and regional scales
(Table 3.2). The national scale analysis was based on the 28 Nomenclature of Territorial
Units for Statistics level III (NUTS III) areas on the Portuguese mainland, the regional
analysis focused on 83 Northern Portuguese municipalities. The latter was selected due to
its diversity in both forest types and socio-economic characteristics. Regional analysis was
also necessary because it was not practical to calculate detailed landscape metrics at a
national scale.

Socio-economic data were obtained primarily from the website of the Portuguese national
statistic office, Instituto Nacional de Estatística (INE, 2001). This information included a
development index for Portugal that quantifies the level of economic and social
development for NUTS III regions and municipalities using 1998 and 1999 data (Fonseca,
2000).

Details of forest conditions and other characteristics at NUTS III level were obtained from
the two most recent Portuguese forestry inventories (1995-1998) and (2005-2006). This
information is available for download from the website of the major forestry institution in
Portugal, Direcção Geral dos Recursos Florestais (DGRF, 2007a). Data at municipal level
were derived from 1:25.000 scale land-use map sheets for 1990 (Carta de Ocupação do
Solo COS‟90) downloadable from the Portuguese Geographic Institute (IGEO, 1990). The
land-use sheets were appended and matched with the boundaries of the municipalities
using union commands in the ArcGIS software. Detailed land cover categories were
combined to produce urban, unproductive, agriculture, water bodies, broadleaved and
coniferous forest classes. In subsequent analysis the broadleaved and coniferous classes
were separated into different tree species namely maritime pine, oak, other broadleaved
trees and eucalyptus.

80
 Table 3.2. Variable descriptions and data sources

Scale Category Variable Description Source Supporting

references

National Socio- UN91 Unemployment rate 1991 INE (Gamborg and

and economic UN98 Unemployment rate 2001 Larsen, 2003)
Regional CPOP Change in population between
1991_ and 2001 (Noss, 1999)
Number of enterprises 1996
N.soc (societies) (Vora, 1997)
Pry_91 Percentage of economic activity
primary sector 1991
Sec_91 Percentage of economic activity
secondary sector 1991
Ter_91 Percentage of economic activity
tertiary sector 1991
DOM91
Domestic electricity consumption
1991

IND91 Industrial electricity

Illi_91 consumption Illiteracy in 1991
NDOC Number of medical doctors/1000
inhabitants

Dev_Id Development Index 98/99 Fonseca

(Included in the ranking but not in 2000
PCA)

Forest Ba80/05 Area burnt in the period DGRF (Moreira et

condition N.F 1980/2005 Number of fires that al., 2001b)
occurred between 1980/2005 IGEO
Brod. Percentage of broadleaved trees
Conif Percentage of Coniferous trees
Forest Percentage of forests

Regional Landscape NP Number of Patches. Equals the COS‟90- (Vora, 1997)

metrics number of patches of the IGEO
corresponding patch type. (Botequilha
Leitao and
PS Equals the sum, across all patches COS‟90- Ahern, 2002)
in the landscape, of the IGEO
corresponding patch metric
values, divided by the total
number of patches.

PD Equals the number of patches of COS‟90-

the corresponding patch type IGEO
divided by total landscape area
(m2) converted to hectares.

PLAND Percentage of landscape COS‟90-

quantifies the proportional IGEO
abundance of each patch type in
the landscape.

3.2.2. Analysis techniques

Landscape metrics were calculated at the regional scale using the FRAGSTATS program
version 3.3. (McGarigal et al., 2002). The run parameters in FRAGSTATS were set with a
81
pixel size of 30 meters and calculations were performed accounting for eight neighbouring
pixels. From the set of metrics proposed by Botequilha Leitão and Ahern (2002) we
calculated class metrics such as percentage of landscape (PLAND), patch size (PS) and
patch density (PD). These metrics were selected on the basis of their ease of interpretation,
anticipated relationships with sustainability (see Table 3.1) and to provide coverage of both
composition and configuration dimensions. Summaries of the metrics and land-use profiles
for each municipality were subsequently exported to the SPSS software for statistical
analysis.

3.2.2.1. Ranking method

The first approach used to compare forest characteristics in different development

situations was a ranking method (Malczewski, 1999; Munda, 2005). Each NUTS III area
was ranked on five socioeconomic development and five forest condition variables (see
Table 3.3). At the regional scale it was also possible to rank each municipality on five
landscape metrics. In all the rankings the lowest value (i.e. 1) was given to the area with
the poorest performance and the highest (i.e. 28 in the case of the NUTS III and 83 in the
case of the municipalities areas) to that with the best. Table 3.3 summarises how the end
points of each ranking scale were defined. Once the individual variables had been ranked,
overall measures for socio-economic development, forest conditions and landscape metrics
were obtained by calculating the average rank for each observation on the different sets of
variables. This meant that there were two final average rankings for each NUTS III area
and three for each municipality. Simple exploratory analyses of the extent to which these
assessments coincided were then performed by dividing the overall measures into two
classes (above and below the median values) and displaying the results on maps or
scatterplots. The degree to which the average rankings corresponded was also assessed by
calculating Spearman rank correlation coefficients.

82
Table 3.3. Characteristics used to rank variables

Category Variable Good Poor

Socio- Increase in Highest value Lowest value

economic population

Development index Highest value Lowest value

Primary sector Lowest values Highest values

Unemployment Lowest values Highest values

Number of Highest values Lowest values

enterprises

Forest Burnt area Smallest area of burnt forests Largest area of burnt forest
condition
Eucalyptus area Smallest area of Eucalyptus Largest area of Eucalyptus

Broadleaved area Largest area of broadleaved Smallest area of broadleaved

trees trees

Coniferous area Largest area of coniferous Smallest area of coniferous

trees trees

Percentage of forest Highest forested area Lowest forested area

in the spatial unit

Class metrics Number of High number of patches Low number of patches

broadleaved patches

Number of High number of patches Low number of patches

coniferous patches

Mean area of High mean patch size Low mean patch size
broadleaved patches

Mean area of High mean patch size Low mean patch size
coniferous patches

Broadleaved patch High patch density Low patch density

density

Note: High patch density of broadleaved tree species was interpreted as “good” due to its
association with the presence of native tree species such as oak.

3.2.2.2. Principal Component Analysis (PCA) and Cluster Analysis (CA)

One limitation of the ranking approach was that it took no account of intercorrelations
between the original input variables. To tackle this and extend the analysis we undertook a
Principal Component Analysis (PCA) and k-means cluster analysis (CA) at the regional
scale. Values for the 83 municipalities on the eleven socio-economic variables in Table 3.2
provided the input to the PCA and the scores on the most important components were then
used to classify the areas. This approach to socio-economic grouping was adopted in
preference to a reliance on the Fonseca (2000) development index because we wanted to
ensure that more account was taken of social characteristics (e.g. population change) as
83
well as economic ones. Nevertheless, to help with the interpretation of the component
scores we also correlated them with the Fonseca (2000) development index.

Following the CA, key characteristics of the groups of municipalities were assessed by
calculating mean values for variables such as population density (MPOP), change in
population (MCPOP), percentage of primary sector activity (MPS) and percentage of
tertiary sector activity (MTS). Analysis of Variance was then used to compare landscape
metrics for three different tree species (eucalyptus, maritime pine and oak) across the
socio-economic groups of municipalities. These tree species were selected to represent the
native broadleaf cover (oak), coniferous plantations (maritime pine) and species introduced
from the 1970s onwards for economic reasons (eucalyptus).

3.3. Results

3.3.1. National scale trends

Figure 3.1 plots the 28 NUTS III areas using their average rankings for forest and socio-
economic conditions. The distribution shows no obvious relationship between the two
variables and the Spearman rank correlation of -0.20 was not significant (p = 0.31). All of
the areas were also classified into four groups according to whether they were above or
below the median values on the two variables.

25.0
Mean rank of 5 forest condition variables

Alentejo Litoral
20.0

15.0

Minho-Lima
Grande Lisboa
10.0

5.0

5.00 10.00 15.00 20.00 25.00

Mean rank of 5 socio-economic variables

Figure 3.1. Mean ranks of Portuguese NUT III areas on forest condition and socio-economic
variables.

84
Figure 3.2 maps the results and indicates some clear geographical blocks with the coastal
areas generally having higher levels of socio-economic development. There was also an
obvious group of areas with below median forest conditions and socio-economic
development in the north east of Portugal.

Figure 3.2. Classification of Portuguese NUT III areas on the basis of forest condition and socio-
economic development.

3.3.2. Regional scale trends

The plot in Figure 3.3 shows a much stronger association between forest and socio-
economic characteristics at regional scale than was apparent in the national analysis. For

85
the 83 municipalities in the Northern region there was a significant negative Spearman
rank correlation of -0.60 (p < 0.01) indicating better forest conditions in the less developed
municipalities. The Ordinary Least Squares (OLS) regression line in Figure 3.3 shows the
general trend, but it is also apparent that there was considerable variation around this (r2 =
33.9%).

At the municipality scale it was also possible to compare an average ranking derived from
five landscape metrics (see Table 3.3) with that for forest conditions. The plot in Figure 3.4
shows a positive association, with a significant Spearman rank correlation of +0.44 (p <
0.01). Compared to Figure 3.3 there is greater variation around the OLS regression line (r2
=
19.1%).

80.00
Mean rank of 5 forest condition variables

70.00

60.00

50.00

40.00

30.00

R Sq Linear = 0.339

20.00

20.00 30.00 40.00 50.00 60.00 70.00 80.00

Mean rank of 5 socio-economic variables

Figure 3.3. Mean ranks of Northern region municipalities on forest condition and socio-economic
variables.

86
 80.00

Mean rank of 5 forest condition variables

70.00

60.00

50.00

40.00

30.00

R Sq Linear = 0.191

20.00

20.00 30.00 40.00 50.00 60.00 70.00 80.00

Mean rank of 5 class metrics

Figure 3.4. Mean ranks of Northern region municipalities on forest condition and class metric
variables.

3.3.2.1. Socio-economic classification

The result of the PCA (with varimax rotation) on the 11 socioeconomic variables indicated
that two components explained 70% of the original variance. Loadings stronger than +/-
0.50 were considered high loadings. Principal Component 1 (PC1) had high loadings for
change in population between 1991 and 2001 (-0.84), percentage of economic activity in
the primary sector (+0.76), illiteracy (+0.93) and rates of domestic and industrial electricity
consumption, -0.93 and -0.66 respectively. Positive scores on this component were
interpreted as an indicator of rurality. PC2 had high loadings on the percentage of
economic activity in secondary (-0.84) and tertiary sectors (+0.89) and the number of
medical doctors per inhabitant (+0.69). The highest positive scores on this component were
for the largest urban centres such as Porto. To further understand the meaning of the PCs
the two sets of scores were correlated with the Fonseca (2000) index of development (not
included in the PCA input). The results were a significant negative correlation of -0.88 (p
<0.01) between the index and PC1 and an insignificant positive association of +0.11 (p =
0.34) with PC2. These results indicate that, as intended, the PCA-based approach covered
broader socio-economic dimensions than the Fonseca (2000) index.

A k-means cluster analysis was used classify the municipalities into five groups on the
basis of their component scores on PC1 and PC2. The five group solution was selected
87
because this was felt to provide the best substantive representation of socio-economic
contrasts within the region. Table 3.4 lists the number of municipalities in each group and
their mean values with respect to population density (MPOP), change in population
(MCPOP), primary (MPS) and tertiary (MTC) sector economic activity. These statistics
indicate some clear contrasts that can be readily interpreted as an urban-rural gradient.
This classification was also different from that obtained using the Fonseca (2000) index
alone where there were very uneven numbers of municipalities across the five groups.

Table 3.4. Characteristics of the five socio-economic groups

Classification Number of MPOP MCPOP MPS MTS

Municipalities
(inhabitants) (%) (%) (%)

Urban centre n=1 5787 -13.0 0.6 88

Inner urban
n=17 959 9.5 1.7 73
fringe

Outer urban
n=17 498 11.2 1.5 54
fringe

Developing
n=12 119 -0.3 5.9 52
rural

Deep rural n=36 52 -10.3 11.2 65

Note: MPOP Mean population density, MCPOP Mean change in Population, MPS Mean Primary
sector, MTS Mean Tertiary sector

Figure 3.5 maps the results of the classification. Porto was identified as the sole urban
centre. The coastal areas around Porto and the most developed municipalities in Northern
Portugal can be described as inner urban fringe. Beyond this there is a similar sized set of
municipalities that can be termed the outer urban fringe. The two remainder groups can be
denoted as developing rural and deep rural. Within the latter there are also a number of
municipalities that contain land designated as National Parks. Tourism is important in
these areas, as reflected in the increase in the MTC percentage in Table 3.4.

88
Figure 3.5. Socio-economic classification of Northern region municipalities.

3.3.2.2 Variations in class metrics

Three class metrics (percentage of landscape, mean patch size, and patch density) for three
tree species (eucalyptus, maritime pine and oak) were compared across the five socio-
economic groups discussed above. These three metrics were included in the set proposed
by Botequiha Leitão and Ahern (2002) and their importance has also been highlighted in
earlier studies such as Forman (1995).

Percentage of landscape (PLAND) quantifies the proportional abundance of each patch

type in the landscape. It approaches 0 when the class type becomes increasingly rare in the
studied area and 100 when the entire area consists of a single patch type (McGarigal et al.,
2002). Figure 3.6 shows how the mean values of PLAND varied across the socio-economic
groups. Maritime pine stands out as the dominant class in all socio-economic categories
while the value for eucalyptus was highest in the outer urban fringe group and that for oak
in the deep rural areas.

89
 14.00

12.00

10.00

Mean PLAND
8.00

(%)
6.00

4.00

2.00

0.00
Urban centre Inner urban fringe Outer urban fringe Developing rural Deep rural
Socio-economic groups

Maritime pine Eucalyptus Oak

Figure 3.6. Variations in percentage of landscape (PLAND) metrics for three tree species across the
socio-economic groups of municipalities.

Figure 3.7 illustrates how patch size (PS) varied across the groups. The mean patch area
equals the sum of the area across all patches in the municipality, divided by the total
number of patches in the municipality (McGarigal et al., 2002). Patch sizes for oaks
increased across the urban-rural gradient, those for maritime pine displayed a more mixed
trend, and the values for eucalyptus were higher in the urban fringe than the rural
categories.

30.00

25.00

20.00
Mean PS
(ha)

15.00

10.00

5.00

0.00
Urban centre Inner urban fringe Outer urban fringe Developing rural Deep rural
Socio-economic groups

Maritime pine Eucalyptus Oak

Figure 3.7. Variations in patch size (PS) metrics for three tree species across the socio-economic
groups of municipalities.

90
Patch density (PD) values for individual tree species are generally interpreted as a measure
of fragmentation (McGarigal et al., 2002). Figure 3.8 indicates that across all five socio-
economic groups the PDs were generally highest for maritime pine, while those for oak
and eucalyptus were similar in several categories. For each tree species it was also the case
that the lowest mean PD occurs in one of the two rural categories.

0.80

0.70

0.60

0.50
(patch/ha)
Mean PD

0.40

0.30

0.20

0.10

0.00
Urban centre Inner urban fringe Outer urban fringe Developing rural Deep rural
Socio-economic groups

Maritime pine Eucalyptus Oak

Figure 3.8. Variations in patch density (PD) metrics for three tree species across the socio-
economic groups of municipalities.

Analysis of variance was used to assess the significance of differences in the class metrics
across the socio-economic groups. Since there was only one urban centre municipality this
category had to be excluded from the analysis and the comparisons in Table 3.5 are based
on the remaining four groups. The results indicate that all three tree species had significant
differences in either PLAND or PD, while none of the contrasts in PS were significant at
the 0.05 level.

Table 3.5. Significance of differences in class metrics across the socio-economic groups.

Tree species Percentage of Patch size Patch density

landscape

Maritime pine 0.01 0.39 0.18

Oak 0.40 0.15 0.03

Eucalyptus 0.01 0.15 0.01

Note: Values in bold are statistically significant at the 0.05 level

91
3.4. Discussion

The first goal of this study was to investigate relationships between forest conditions and
measures of socio-economic development at national and regional scales in Portugal. At
the national scale there was only a weak correlation between socio-economic development
and forest conditions in NUT III areas. However, a stronger and statistically significant
negative correlation was identified at the regional scale.

A weaker association at the national scale can be attributed to the interaction of several
factors. As Figure 3.2 indicates, there was a clear tendency for poorer forest conditions to
occur in a number of the most urbanised areas (e.g. around Porto, Lisboa and on the
Algarve). In rural regions the situation is more variable, with a particular contrast in forest
conditions between the inland parts of southern Portugal (e.g. the plains of Alentejo) and
the more mountainous area further north. Alentejo is dominated by a livestock-based agro-
forestry system known as “montado” with scattered cork trees which are relatively easy to
maintain in good condition (Firmino, 1999). Forests further north are characterised by
relatively dense stands of pines, eucalyptus or oaks which are more challenging to manage
and vulnerable to wildfires. With hindsight, variables such as the percentage of
broadleaved trees used in the national analysis (although the best available) were not
sufficiently sensitive to these regional differences in tree species or forestry systems. In
addition, the absence of landscape metrics meant that factors such as contrasts in stand
structure were not taken into account. Further analysis could be conducted to tackle these
problems, but the data and processing requirements (e.g. to generate landscape metrics at a
national scale) would be considerable.

Almost all of the Northern region included in the municipality analysis had below median
forest conditions on the national scale (see Figure 3.2). However, restricting the analysis to
this region helped to control for some of the key differences in tree species and forestry
systems. The outcome was a stronger trend for higher levels of socio-economic
development (mainly near the coast) to coincide with poorer forest conditions (Figure 3.3).

A second objective was to examine whether there was a systematic variation in forest
landscape metrics according to the stage of socio-economic development. The initial
ranking analysis at the regional scale indicated a positive association between measures of
forest condition and a set of landscape metrics (Figure 3.4). More detailed assessment of
differences in specific metrics for particular tree species was then conducted. The results
suggested a number of trends and contrasts (see Figures 3.6 to 3.8 and Table 3.5),

92
supporting the conclusion of Botequilha Leitão and Ahern (2002) that metrics such as
PLAND, PS and PD are of value in evaluating sustainability issues.

Interpretation of the differences in the tree species metrics can be facilitated by reference to
the forest transition concept (Mather, 1992). In essence, there has been a transformation of
natural Portuguese broadleaf (particularly oak) forest into more productive tree species
alongside the broader process of socio-economic development. The deep rural category of
municipalities still has the largest proportion (2.5%) of its area with oak trees. By contrast,
the outer urban fringe group have much higher percentages of landscape with non-native
species such as eucalyptus (Figure 3.6). The patch sizes for pine and eucalyptus species are
also much larger (Figure 3.7), often being over 15-20 hectares. Both these characteristics
reflect planned initiatives (e.g. state funded development of coniferous plantations) in
municipalities such as Bragança, Chaves, Viana do Castelo and Vila Real during the 1940s
(Brouwer, 1999; Roche, 1998) and more generally these areas have features typical of
“industrial” forestry.

The presence of larger patches and higher percentages of area occupied by the same tree
species would, at first sight, suggest a connected landscape. However, the interpretation of
the patch density metric (Figure 3.8) shows a more fragmented landscape in urbanised
areas. This is a common trend and reflects a number of other studies (Antrop, 2004;
Forman, 1995).

The values of patch size show some signs of increasing across the urban-rural gradient
(Figure 3.7), but the differences were not statistically significant (Table 3.5) and the
change is not as marked as might be anticipated in some cases (e.g. for oaks). Several
factors help to explain this result. One is that the deep rural group (where forests might be
expected to cover larger areas in bigger patches) includes some mountainous regions where
the environmental conditions are unfavourable for large areas of trees. A second factor is
the presence of a number of National Parks with distinct features such as absence of
eucalyptus trees and an emphasis on tourism i.e. “post-industrial” forestry.

Taken together, these results suggest that the different socio-economic groups have
contrasting forest characteristics which reflect the pre-industrial, industrial and post-
industrial categories embedded in the forest transition concept. It is also worth noting that
the metrics used in this study are quite effective in distinguishing the “industrial” from
other categories of forestry, but less so in separating pre and post industrial features.
Within the deep rural group there are still common lands where forests provide a variety of
products characteristic of pre-industrial forestry (Brouwer, 1999), as well as protected

93
areas where the service functions characteristic of post-industrial forestry are apparent. In
addition, although there has been a reversal of forest loss in many developed areas, the
characteristics of these new forests are not especially appropriate to fulfil the increasing
needs of recreation and tourism activities. This highlights that it is important to consider
issues of functions as well as composition/configuration in assessing issues of landscapes
and sustainability (Blaschke, 2006; Perz, 2007).

The third goal of this study was to assess the extent to which landscape metrics can inform
strategies to enhance forest sustainability. From the previous discussion it is clear that
landscape metrics can be used to identify some of the different problems and issues in
urban and rural areas. However, it is also evident that they have some limitations in terms
of distinguishing some of different functions that forests can perform. This highlights that
recommendations to improve forest condition need to be place specific and take account of
other surrounding land uses (since these will influence the functions that forests need to
perform). In other words, forests need to be better integrated into the continuum of land
uses that encompasses landscapes.

Planning strategies have been defined by Ahern (1995:139) as protective, defensive,

offensive and opportunistic. Acknowledging such a view, a key protective step in the deep
rural areas would be an increase in both the percentage of landscape and patch size of oaks
and other broadleaved trees. This would help tackle a critical problem in these areas,
namely the disappearance of the “idyllic landscape” shaped by “traditional farmers” with a
mixture of low intensity crop, grazing and forest land uses (Firmino, 1999). Enlarging the
area and patch sizes of broadleaves would help to support economic and social
regeneration through other means (e.g. tourism), but needs to be accompanied by measures
to maintain other, potentially conflicting, activities such as livestock grazing. A challenge
for management of forests in these rural areas is thus to create adjacent uses with low level
of conflict. Contagion/interspersion metrics could play a role in helping to identify
compatible land use mosaics that would provide sustainability benefits in such areas.

By contrast, according to Ahern (1995) the most appropriate strategy for an urban centre is
a defensive one. Consequently, the priority in urban areas should be to enhance landscape
connectivity in order assure provision of environmental goods and services. To achieve
this it seems sensible to propose an increase in both percentage of landscape and patch size
of oaks and other broadleaved trees. This is a similar recommendation to rural areas, but
the silvicultural techniques (e.g. tree spacing and stand age composition) would need to
vary according to management priorities (i.e. to connect the landscapes in urban areas and
promote economic and social regeneration in rural areas). Thus, in urban areas
94
management priorities should focus on enhancing vegetation structure in existing
woodlands as well as improving connectivity between them. These recommendations are
similar those presented by Hedblom and Soderstrom (2008) in their study of urban
woodlands in Sweden. They also note that many of the challenges to forest management in
urban areas are concerned with reconciling “management for biodiversity” (Hedblom and
Soderstrom, 2008) and aesthetics, for instance dealing with the fact that dead wood both
“enhances biodiversity” and is “not aesthetically attractive”. Diversity and connectivity
measures are therefore examples of the types of landscape metrics that can help in planning
enhancements to the sustainability benefits associated with forests in urban areas.

3.5. Conclusions

The approach presented in this study is innovative in correlating ecological landscape

metrics for tree species with socio-economic indicators for the areas in which the forests
occur. It therefore provides an example of a simple interdisciplinary method for
investigating sustainability issues and demonstrates that landscape metrics can be used as a
measure of socio-economic change.

The results of this study have several implications for the wider landscapes and
sustainability debate. Our study suggests that it is possible to identify a positive association
between landscape metrics and measures of better condition for forests and that the
calculation of metrics for individual tree species can help inform strategies for sustainable
forest management in urban and rural areas. The research also demonstrates how forest
characteristics vary over an urban-rural gradient and therefore reinforces the point that
criteria for sustainable management or strategies to establish “virtuous circles” need to be
context specific (Matthews and Selman, 2006).

Landscape metrics clearly have value in detecting patterns of land cover or environmental
change, but those used in this study are limited in their ability to distinguish the different
functions or services of forests. This capability is an important aspect in any assessment of
sustainability at the landscape scale (Selman, 2006) and it therefore suggests that future
research should either evaluate more refined metrics to represent such functions or
examine how basic metrics can be best supplemented by other indicators in the toolbox for
sustainable landscape management.

95
Chapter 4. Integrating public uses and preferences in the
design of multifunctional plans at the catchment scale: A
case study in the Minho region of Portugal

96
Abstract

This chapter explores the potential role of forests in sustainable landscape management
within two surface water catchments located in the Minho region of Portugal. In order to
do so, the characteristics of the forests were surveyed by gathering data such as stand
composition and diameter classes. Subsequently, a questionnaire survey collected data
concerning public uses and preferences for forests. Combining a literature review about the
roles of forest in the area with both the field and the questionnaire surveys it was possible
to distinguish between the role(s) of forests across the area of two watersheds. Although
the analysis of the public uses and public preferences for management strategies in each
municipality revealed the need to implement multifunctional forests, the results also
indicate that there is a need to vary the type of multifunctionality within the area of the two
watersheds. In deep rural areas an integrated type of multifunctionality is critical due to the
need to fully integrate protection, recreation and production functions. On the contrary, in
more urbanised areas dominant uses such as recreation and production are of foremost
importance indicating that the option of spatially separating the functions (spatial
multifunctionality or dominant use) is likely to be a wiser strategy as it works in tandem
with the dynamics currently in place. However, in order to implement those strategies, as is
implicit in the Water Framework Directive (WFD), it is crucial to involve all forest
stakeholders in the creation of whole catchment governance strategies. The results show
that it is likely that sustainable urbanized landscapes will imply completely different
aspects than sustainable rural ones. However, it is also likely that rural and urban
sustainabilities will differ from each other. As a consequence, it is suggested that there
need to be strategies to strengthen rural/urban patterns of development based on an
ecosystem approach to watershed management. In order to do so it is crucial to i) integrate
forestry within a formal planning context ii) create public-private partnerships to address
the multitude of ecosystem services forests have the capacity to provide iii) transcend
property boundaries to implement desirable forest management strategies. Based on a set
of simple tools such as questionnaire and field surveys supported by Geographic
Information System (GIS) techniques the study highlights ways to move towards
sustainable forestry management in two watersheds in Northern Portugal. Further, the
results suggest that even in municipalities within similar characteristics (either urban or
rural) it is likely that the role(s) of forests may be different. As so, further research
addressing smaller scales (e.g. parish) are likely to be important to further exploring the
role(s) of forests within the study area.

97
4.1. Introduction

Sustainable landscape planning is concerned with guiding the actions and interactions of
natural and human factors so landscapes lose neither their ecological integrity nor their
ability to fulfil basic human needs (Brandt et al., 2000; Fry, 2001; Kozlowski and Hill,
1993; Mander et al., 2007; Matthews and Selman, 2006). Sustainable landscapes are
inherently multi-functional linking people and their natural environment (Brandt et al.,
2000; Fry, 2001; Mander et al., 2007). According to the European Landscape Convention,
planning is “a forward-looking action to enhance, restore or create landscapes” (Council of
Europe, 2000). In addition to effective planning, there is a need to put into practice
effective governance systems for sustainability (Dolman et al., 2001; ELC, 2000; OECD,
2006). Governance over a territory can be understood as “the emergence and
implementation of innovative shared forms of planning and managing of socio-spatial
dynamics” (CEMAT, 2007:29). Thus, sustainable landscape governance requires the
negotiation of a set of commonly agreed objectives and a framework of shared
responsibilities to effectively enable spatial development strategies and policies (CEMAT,
2007).

During the late 20th century landscape functions (e.g. nature conservation, leisure) have
tended to become segregated in most European landscapes as a result of the specialisation
and intensification of production (Mander et al., 2007). Selman (2006:15) considers this
functional separation of land uses to be an underlying contributor to many environmental
problems. On the contrary, if a land use type is able to provide multiple uses and functions
it is more likely that it will fulfil the needs of a broader group of people (Baskent, 2007;
Baskent et al., 2000). Consequently, multifunctionality either from a single land use type
such as agriculture (multifunctional agriculture) or from several land use types
(multifunctional land use) is increasingly being promoted as a means of moving towards
sustainability (Fry, 2001; Selman, 2002). Furthermore, multifunctionality has increasingly
been proposed as a principal “hallmark” of landscape strengthening its case for being at the
heart rather than the periphery of integrated spatial planning (Hector and Bagchi, 2007;
Selman, 2006).

Multifunctional forestry can be achieved in two distinct ways by, 1) pursuing different
goals in a corresponding mixture of separate land use types or 2) integrating consistently
different goals from the beginning in order to accomplish them simultaneously (de Blust
and Olmen, 2000). The first way can be defined as spatial multifunctionality because
different spatial units have an unequivocal goal and management. In spatial

98
multifunctionality each piece of land has one function and thus when zooming in or out an
area appears to be more or less multifunctional. By contrast, in the second case different
goals are attained in the same spatial unit. This type of multifunctionality can be defined as
integrated multifunctionality (de Blust and Olmen, 2000).

Different users and managers of natural resources (e.g. foresters, livestock grazers) as well
as governments shape the landscape in order to achieve different outputs from landscapes
such as, productivity, diversity, integrity and stability, essentially giving landscapes a
multifunctional character (Tress and Tress, 2001). However, a range of settings, from
urban to rural, find their multifunctionality and distinctiveness compromised (Selman,
2002). It is acknowledged that some “cultural” landscapes in post-industrial societies are
not self sustainable because the links between landscapes, community and economy are no
longer self reinforcing (Selman, 2006). As a consequence there are multifunctional
landscapes where sustainability is weak. In these cases, there is a need to (re) create
“viable” multifunctional land use (Pinto-Correia and Breman, 2008).

Recent work has suggested that sustainable development strategies should focus on
encouraging „„virtuous circles‟‟ in landscapes (Matthews and Selman, 2006) so that the
linkages between the socio-economic sphere and environmental functions are reinforced.
Also, Haines-Young (2002), based on the concept of natural capital proposes the
“landscapes and sustainability model” for addressing the sustainable landscapes challenge.
In this model it is argued that landscapes are more or less sustainable in terms of the
outputs of goods and services that are important to people in a specific context. This
model was criticized by Blaschke (2006) because it mistreats issues related to the spatial
arrangement of natural resources that certainly influence the provision of goods and
services (Hersperger and Forman, 2003). The combination of both arguments
(goods/services and spatial configuration) into a more holistic approach, is more likely to
address these major issues in planning for sustainability by acknowledging socio-economic
and environmental distinctiveness which may be able to reinforce different processes
across landscapes (Baskent and Yolasigmaz, 2000)

Forests play a crucial role in conserving biological diversity, water resources, soil and the
maintenance of ecological functions and ecosystems integrity (Sayer and Maginnis, 2005).
Despite agreement about the important role of forests in sustainable development (Sayer
and Magginnis, 2005; Slee, 2006) there are problems when it comes to the implementation
of sustainable forestry management (SFM) across a range of landscape types (Elands et al.,
2004; Elands and Praestholm, 2008). Forests are normally studied either in a rural (Elands
et al., 2004; Elands and Praestholm, 2008; Slee and Snowdon, 1999) or urban context
99
(Matsuoka and Kaplan, 2008), but the ways in which forests may help to reinforce the
links between the urban-rural are not obvious in the literature. Consequently, it is
important to study the ways in which forests may facilitate the creation of “virtuous
circles” across urban-rural regions.

Trees, both inside and outside woodlands and forests, are increasingly being recognized for
their important contribution to humanity through the provision of fuel, fibre, habitat for
biodiversity, carbon storage and amenity (Sayer and Magginnis, 2005). Sustainable
forestry management (SFM) and the ecosystem approach (EA) as applied to forests address
the critical contribution forestry makes to broader sustainability issues. However, both
SFM and EA do not address the issue of how ecosystem functions such as production,
recreation and water and soil protection can be supported across bio-geographical areas
such as river basins (Sayer and Maginnis, 2005). This has become more important recently
as forestry moves from a primarily production based focus to include consumptive (e.g.
recreation activities in aesthetically pleasing forest areas) and protective (e.g. providing
biodiversity, avoid floods and soil erosion) services. As a consequence, mechanisms for
the reconciliation of multiple functions across landscapes are of critical importance (Elands
and Praestholm, 2008; Elands and Wiersum, 2001; Selman, 2006).

The Understanding Forestry In Rural Development research project (UFIRD) identifies

four major ways in which forests may contribute to sustainable development in rural areas,
by providing: 1) direct marketable goods, 2) “shadow” or “halo” benefits, 3) non market
goods/services and finally by 4) promoting local identity based on culture and history
(Slee, 2006; Slee et al., 2004). Table 4.1. shows the forest contribution, the goods and
services generated, the methods used for its estimation as well as the functions of forests
that are associated with the provision of these benefits.

100
Table 4.1. Goods and services provided by forests and estimation methods

Forest Description Goods/services Estimation methods Function of

contribution provided forest

Conventional The impact of forestry activity, Timber, Surveys within Production

marketable including forest related work forestry sector
Non-timber
goods and the upstream and followed by
products (e.g.
downstream connections of Keynesian local
honey),
forestry on employment and income, employment
employment,
income; multipliers
income

“Shadow “or The indirect impact of forestry Income Focus groups Protection,
generated due broader group of Recreation
“halo” non on surrounding economic
to presence of sectors followed by
marktable activity forests (e.g. Keynesian local
increase in income, employment
goods/services
value of multipliers
property)

Conventional The non-market values of Carbon Field surveys Protection

non forests and woodland, sequestration, characteristics of (also
marketable biodiversity, forests followed by productive
which although not generating
goods air quality, benefit transfer e.g. carbon
immediate regional
recreation methods credits)
income, do create a contribution
to national green accounts

Social (non The social values attributable to Community Focus groups local Recreation,
forests and woodlands which identity, /broader group of
marketable
range from their contribution to culture, sectors followed by
goods) symbolic capital and traditions interpretative
community identity to their methods
contribution to social capital
building.

Source: Slee (2006)

In a high diversified Europe Elands and Wiersum (2001), distinguish five different roles
for forestry; agri-ruralist, hedonist, utilitarian, community stability and nature
conservation. Within each of these five rural development discourses forests may have
distinct roles. These roles vary from providing income and employment (in the utilitarian
discourse), complementing and diversifying farming systems (agri-ruralist), to supporting
socio-economic regeneration of marginalized areas by improving living conditions in
remote places (community stability), as well as assuring biodiversity (nature conservation).
Therefore, multifunctional forest management can overcome some of the problems that
rural landscapes face (e.g. source of direct income or through recreational activities).
However, the appropriateness of these discourses in different locations within a watershed
catchment has not been explored.

101
In urban areas, forests are well known for the provision of protective environmental
services (Table 4.2) such as air filtering, regulating city temperatures, noise reduction
recreational provision (Matsuoka and Kaplan, 2008). Furthermore, forests may have
distinct roles in assuring the welfare of urban and rural residents.

Table 4.2. Services of forests in urban areas

Service Description and justification Sources

Air filtering Vegetation reduces air pollution but to what level seems to depend on the (Bolund and
local situation. Because of the larger total surfaces of needles, coniferous Hunhammar,
trees have a larger filter capacity than the trees with deciduous leaves. This 1999)
capacity is also grater because the needles are also shed during the winter.
But coniferous trees are sensitive to air pollution and deciduous trees are
better to absorb gases.

Micro- Local climate and even weather is affected by the city. This phenomenon, (Bolund and
climate called urban heat island effect, is caused by the large area of heat Hunhammar,
regulation, absorbing surfaces, in combination with amounts of energy use in cities. 1999)
at street and All natural systems in urban areas will help to reduce these differences.
city level

Rainwater In vegetated areas only 5-15% of the rain water runs off the ground, with (Bolund and
drainage the rest evaporating or infiltrating the ground. In vegetation free cities Hunhammar,
about 60% of the rainwater is instead led off through storm water drains. 1999)

Sewage Taking care of sewage costs cities large amounts of money, and the (Bolund and
treatment nutrients that are still released contribute to eutrophication of the Hunhammar,
surrounding water ecosystems. Some studies have been showing that 1999; Johnson
wetlands and forests can significantly reduce the costs with sewage et al., 2002)
treatments

Recreational The recreational aspects of urban ecosystems are perhaps the most valued (Brainard et al.,
and cultural ecosystems services in the cities 2001; Matsuoka
values and Kaplan,
2008)

However, there is a need to go beyond distinguishing between different roles for forestry in
rural and urban and explore the ways in which rural/ urban partnerships may be reinforced
(Antrop, 2005, 2006; Wu, 2006). This can be achieved by investigating the ways in which
the integration of forest functions (production, protection and recreation) can be made
across the entire area of catchments comprising urban and rural areas (Sabatier et al.,
2005). Forests provide different functions depending their location, with both “in situ” as
well distant locational effects either omnidirectionally or according to the flow of a river
(Fisher et al., 2004). These different functions can be provided through monofunctional or
multifunctional land use in forest areas (Stevens and Montegomerey, 2002).

The forestry sector has had a long tradition of thinking in “multiple uses” and “multi-
purpose” being these recurrent terms in the forestry literature (Stevens and Montegomerey,

102
2002). Integration of functions in the forestry context has been the focus of many research
studies addressing compatibilities and incompatibilities of forest functions such as timber
and recreation, timber and water quality, timber and wildlife. An overview of these studies
shows compatibility or competition between timber and most other resources (e.g. water
quality) but only rarely incompatibility (Stevens and Montegomerey, 2002). Compatibility
between forest uses is more easily achieved at the regional or forest scale than at the
smaller scales of stand or management unity reinforcing the argument for planning forests
at larger spatial scales (Wu, 2006). Nevertheless, within this literature authors can be found
to support the integration of functions while others advocate spatial multifunctionality or
even dominant use (Stevens and Montegomerey, 2002).

In the area of a single watershed the biophysical relationships between forests, people and
water are highly variable depending on climate, soils and vegetation types (Sabatier et al.,
2005). However, Johnson et al (2002) summarized four simplified basic relationships as
follows:

1. Forests slow the rate of runoff in a watershed. Forest vegetation takes up water and
delays the time to soil saturation. Forest soils usually have a higher water storage capacity
than non forest soils. Furthermore, the more complex structure of the forest ground surface
and underlying soil allows more efficient soil infiltration compared to a deforestated
watershed. By slowing down the rate of runoff, forests can help to minimize flooding in
smaller watersheds. By slowing the runoff rate forests may also increase minimum stream
flows during the dry season.

2. Forest reduces soil erosion and sedimentation of waterways. Interception of rain and
snowfall by forest canopies means that less water falls on the ground compared to a
deforested watershed. Extensive root systems help hold soil more firmly in place and resist
landslides.

3. Forest soils filter contaminants and influence water chemistry. Forest soils are more
waterlogged than other soils (except wetlands) and contain more nutrients, allowing them
to filter out contaminants. For example, streams in agricultural areas in temperate regions
typically have nitrate levels 10 times higher than streams in nearby forested watersheds.

4. Forests reduce the total annual water flow in a watershed. Contrary to popular
opinion forests generally reduce the total annual stream flow. This is because trees
consume water for transpiration which is then evaporated back to the atmosphere. The
degree to which forest reduce stream flow, however, depends on factors such as type of
roots as well as rotation period. For example, shallow-rooted trees tend to use less water
103
than deep rooted trees. Young regenerating forests tend to use much more water than
mature old growth forests.

The hydrological services of forests are amongst the most valuable of the many ecosystem
services from forests (Johnson et al., 2002; Sabatier et al., 2005). As a result, there is a
growing interest in ecosystem based approaches to ensuring clean water services that go
beyond the more typical regulatory measures or subsidies to provide new financial
mechanisms which encourage new forms of land management techniques in private lands
(Johnson et al., 2002; Sabatier et al., 2005). Milligan et al (2009) suggested that the present
moment is appropriate for considering new governance arrangements that adopt and
implement fresh approaches to participatory decision-making and natural resources
management. Accordingly, several key recommendations have been identified for
engaging forest stakeholders to improve participation and effective dialogue around
sustainable forest management.

Worldwide, there are growing examples of different types of financial mechanisms being
used in ecosystem based watershed management (Johnson et al., 2002; Sabatier et al.,
2005). For example, by investing approximately $ 1 billion dollars in land protection and
conservation practices New York City hopes to avoid spending $4-6 billion on end of pipe
filtration and treatment. Also in the US it has been found that every $1 invested in
watershed protection can save from $7.5 to nearly $200 in costs for new filtration and
water treatment facilities (Johnson et al., 2002). According to Johnson et al. (2002) there
are three major types of financial mechanisms for watershed management; self organised
private deals; trading schemes and public payment schemes (Johnson et al., 2002). Self
organized private deals arise where private entities have developed their own mechanisms
to pay for watershed protection with little or no government involvement (e.g. the Perrier-
Vittel group of water bottler in France). The latter two mechanisms imply some form of
governmental intervention either as a regulator of trading schemes or by public payment
schemes. A review of these financial mechanisms and, the situations in which they are
appropriate is beyond the scope of this Chapter (but see Johnson et al. 2002 for a review).
Here, it is important to recognise them as strategies already in practice which support the
argument for promoting sustainable forestry management across regions within the same
catchment, as well as to indicate the important role governments may have in leading this
trend (Johnson et al., 2002).

The European Water Framework Directive (WFD) (Directive 2000/60/EC) is a legally

binding document that requires European member states to implement management
measures to achieve “good water quality” by 2015. The WFD is built upon two main
104
innovative approaches; it proposes on the one hand more integrated ecological definitions
of water and, on the other hand, it reinforces the need for public participation in policy
implementation (Steyaert and Ollivier, 2007). If there is agreement upon the former
approach (ecological benefits of the WFD) as it aims at enhancing the ecological and
chemical status of water bodies, there is still much debate as to how the latter (governance
mechanisms and public participation) can be implemented in a fair manner (Bateman et al.,
2006). One of the problems which WFD aims to tackle is the diffuse emission of pollutants
from agriculture and this may imply drastic changes in already fragile rural economies in
countries such as England (Bateman et al., 2006).

It follows from the above, that forests may have an important role to play in the
implementation of the WFD. For example, forests may filter contaminants and influence
water chemistry. Additionally forests may have a major role in both rural and urban
development processes (Tables 4.1 and 4.2). Despite legally binding documents at the EU
level, such as the WFD which may also be met by sustainable forest management
practices, few countries have mandatory plans for forestry (Selman, 1997). Consequently,
more sophisticated guidance from planning and government agencies is required in order
to best integrate functions in the forestry arena. According to the main steps and deadlines
written into the WFD, detailed river basin plans should be published in 2009 providing an
opportunity to both engage the public and private sectors (including forest stakeholders) in
whole catchment management (INAG, 2005).

This study explores multifunctional forest management using two river catchments in
Northern Portugal as case studies, with the aim of informing future strategies for
sustainable forestry management. This aim was guided by three key questions which are
presented in Box 4.3.

Box 4.3. Questions

1. What role(s) should forests serve in the study area?

2. Is the forest role identical across the area (e.g. in urban and rural areas) of two
watershed catchments?
3. How can forests contribute to the sustainable development of the whole region?

4.2. Data Collection and Methods

4.2.1. Study area

This research explored the contribution of forests to sustainable development by analysing

the potential role(s) of forests within two river basin catchments in the Minho region of
105
northern Portugal. In Chapter 3 it was shown that the level of socio-economic development
of the Northern Portuguese municipalities ranges from deep rural to urban centre
categories (Carvalho-Ribeiro and Lovett, 2009) and were also variation in forest patch size
across the rural-urban gradient. According to the previous study, the patches of Portuguese
pristine forest (oak) had relatively low mean patch sizes (around 5 ha) although higher
mean patch sizes were found in rural areas.

A recent study by Pinto Correia and Breman (2008:1) identified different “vocations” for
the Portuguese municipalities in terms of the role that farming could have in their future.
The authors assumption was that as the territories are different then the type of
multifunctionality municipalities‟ should deliver should also vary. Figure 4.1 shows both
the classification made in the Chapter 3 as well as the one by Pinto Correia and Breman
(2008). The purpose of the two classifications was different and they consequently show
considerable variations. Despite differences, what both classifications indicate is that there
is a need to diversify the type of management strategy for agriculture and forests (e.g for
production, protection and recreation) across the area. As can be seen in the final typology
from Pinto-Correia and Breman (upper map Figure 4.1) the areas in black (1b) were
classified as areas of production and specialised agriculture with high profitability. Around
these municipalities there is a second set that were classified as areas for extensive
agriculture with high environmental quality in diversified landscapes (2b in the legend). In
the top of the catchments one municipality (Ponte da Barca) was included in type 2a in
which the characteristic feature is the extensive silvo-pastoral system. The remaining two
municipalities in the top of the catchment (Terras do Bouro and Arcos de Valdevez) were
classified as 3b. In type 3b, agriculture is clearly residual both in terms of land cover and
economic activity. In these municipalities forest cover is dominant as are extensive grazing
areas, but in contrast to type 2b, the landscape pattern has been progressively simplified as
farming areas disappeared or were reduced in size (Pinto-Correia and Breman, 2008).

106
Figure 4.1. Classification of the area based in two studies

4.2.2. Overall method and approach

Two key methods employed were forest field survey and stakeholder questionnaires. The
field survey was not intended to be a full forest inventory but rather a simple
characterization of the basic stand structure and the cleanliness of the plots, located in six
municipalities spread throughout the catchment (Figure 4.2). The questionnaire survey was
designed to be short to increase the sample size, eliciting stakeholder preferences and
opinions on forest management, use and the provision of ecosystem services. Figure 4.2
107
shows the geographical location of the municipalities and distinguishes between those
where both questionnaire and field survey were conducted (dark shading) and those in
which only questionnaire surveys was undertaken (no shade).

Figure 4.2. Data collection across the municipalities

Figure 4.3 shows the land use of the area using a map from the 90s (IGEOE, COS90). As
can be seen urban and agriculture areas are predominant in the bottom of the catchments
while at the top forests are predominant (Figure 4.3). The Lima and Cávado river basins
flow from a mountainous region in which the highest peak is 1530 meters high to the sea
coast along a distance of approximately 80 kilometres.

108
Figure 4.3. Land use in the study area based on COS 90

4.2.3. Characterisation of forest plots in the study area

Between September and October 2006 a field survey was conducted in six municipalities
of the Lima and Cávado catchments in order to characterise the forests in the study area.
The six municipalities where selected based on their position in the catchment (Figure 4.2).
It was important to survey municipalities in the bottom (Esposende and Viana do Castelo),
at an intermediate location (Ponte de Lima and Braga) and at the top of the catchment (the
mountainous areas of Ponte da Barca and Terras do Bouro). Another important factor in
the selection of the municipalities was formal permission from the municipal forestry
office to undertake the field survey. Within the area of each municipality a random sample
of approximately ten plots were inventoried for characteristics such as stand structure and
other parameters according with the field sheet presented in Appendix 2. The locations of
the plots surveyed were selected by expert staff of the municipal forestry office based on
109
the representativeness of the plot for the municipality. 500m2 circular plots were
established and all the trees species were identified and counted within them, their
diameters at breast height (dbh) were measured and the presence of natural regeneration
and litter were recorded (see Appendix 2 for further details and a copy of the recording
sheet).

4.2.4. Questionnaire survey

The survey was undertaken with help of one field assistant between March and August
2007. The aim was to collect information from around 30 people in each municipality by
interviewing approximately fifteen “forest related people” as well as fifteen members of
the “general public”. Due to time constraints this was not achieved in some municipalities,
however 375 questionnaires were completed. In order to get data from such a broad group
of people, in each municipality a meeting with the municipal forestry office (Gabinete
Tecnico Florestal-GTF) was prearranged in order to gather contacts for forest related
people to be sampled. From this set of contacts a “snow-ball” sample strategy was used to
increase the sample size by asking respondents for the contact details of other users in their
parish or municipality. The general public questionnaires were done as a street survey in
the bigger villages (or cities) in each municipality.

The questionnaire took approximately fifteen minutes to administer and was organised in
four sections: 1) introduction, 2) uses of forests, 3) perceptions of a sustainable forest and
4) characterization of the interviewees (see Appendix 3 for a copy of the questionnaire). In
the introduction the purpose of the study was explained and respondents asked about where
they lived. The second section gathered data relating to the frequency interviewees‟ use of
forests based on several predefined categories, such as recreation and timber collection.
Other questions focussed on the importance that respondents gave to the goods and
services provided by forests in the area they lived at present and what they perceived
would be important in 20 years time. In section three, respondents were asked to rank their
preferred management strategy (production, protection and recreation) for where they
lived. These management strategies were illustrated in the form of photos (Box 4.4).
Interviewees were then asked to select one, from two contrasting characteristics (e.g. pure
versus mixed stands), that would most adequately support the management strategy
selected in the previous question. The two major trends of change were also ranked.
Finally, socio-economic data such as age class, level of education, landownership,
employment and income was collected.

110
Box 4.4. Pictures of forests under different management

4.3. Results
PRODUCTION PROTECTION RECREATION

4.3.1. Characterisation of forest plots in the study area

In six municipalities, 55 plots of 500 m2 were surveyed. From the total of 55 plots, 29 plots
were pure stands of either of broadleaves and coniferous (52%) while 27 had a mixed
composition with both coniferous and broadleaves (49%) these plots having a recurrent
presence of eucalyptus trees. In the vast majority of the plots 80 % (44 out of 55) natural
regeneration was found. Litter was present in 40% of the plots (23 out of 55) revealing
concerns about sanitary condition of the forest plots. Table 4.5 summarises the data
gathered throughout the field survey by municipality, and also indicates how the
municipalities were classified by both Carvalho Ribeiro and Lovett (2009) and Pinto-
Correia and Breman (2008).

111
Table 4.5. Forest characteristics in the surveyed plots

Municipalities N Natural Mixed Diameter Plots CR&L PC&B

regeneration stands (%) > 47.5cm surveyed with
(%) (%) litter (%)

Terras de 10 90.0 45.0 13.0 40.0 Deep Rural 3b

Bouro

Ponte da 10 70.0 40.0 4.3 30.0 Deep Rural 2a

Barca

Ponte de 9 66.7 44.4 1.4 55.6 Developing 2b

Lima rural

Esposende 10 70.0 30.0 15.6 50.0 Outer urban 1

Viana do 9 88.9 37.8 13.9 33.3 Inner urban 2b

Castelo

Braga 8 87.5 30.0 8.9 37.5 Inner urban 3c

CR&L- Carvalho Ribeiro and Lovett (2009), PC&B- Pinto Correia and Breman (2008): 1 Production-
Specialized agriculture with high profitability 2aExtensive agriculture environmental quality homogeneous
landscapes,2bExtensive agriculture environmental quality diversified landscapes 3b Agriculture
environmental quality in forested areas, 3cAgriculture environmental quality in urban areas

The percentage of the plots surveyed in which natural regeneration was present was, in
general, high in all the municipalities indicating a good capacity of forests to regenerate
naturally in the region (Table 4.6). Despite indications that forests in the region regenerate
naturally less than 50 % of the plots surveyed in each municipality had a mixed stand
composition meaning that pure stands were more prominent. The dominant diameter
classes were the smallest of 5, 15 and 25 cm. The 5 cm class included all the trees in which
the diameter at breast height varied between 2.5 cm to 7.5 cm. The 15 cm class included all
the trees in which the diameter at breast height was above 7.5 cm and below 12.5 cm, and
so on. This result indicated that old growth forestry is minimal in the study area. The
geographical distribution of the diameter classes found across the area is shown in Figure
4.4. As can be seen in Figure 4.4 and in Table 4.6 the presence of old growth forest
(diameters above 47.5 cm) is very low in the municipalities surveyed. Also, the presence
of litter (plastic bags, cans) scattered in the plots indicated that more effectively enforced
management of the forests in the region is required.

112
Figure 4.4. Mean diameter class in all plots by municipality

The low number of sampling sites discouraged a statistical comparison of the forests across
the socio-economic groups presented in Chapter 3. However, as shown in Table 4.6, there
were no obvious differences in the conditions of the forests across the groups of Carvalho
Ribeiro and Lovett (2009) or Pinto Correia and Breman (2008). Above all, the field survey
demonstrated that forests in the area are far from demonstrating sustainable forestry
management. This argument is further developed in section 4.3.3.

4. 3.2. Results of the questionnaire survey

From street survey and the meetings with “forest-connected people” the total sample size
collected was 375 questionnaires. From this sample, 57% of the interviewees were male
(213 out of 375) the remaining 162 (43%) were female.

4.3.2.1. Uses of forests

In Minho region of Portugal people use forests in order to get multiple goods and services.
In addition to being used for collection of timber and non-timber products, forests are also
used as recreation sites; traditional activities in forest areas such as grazing and hunting are
still important activities in some locations. The overall data shows that the activities in
which there were higher frequencies of use were recreation, timber collection and grazing.
In general, people that own forest land (landowners) had higher frequencies of use.

113
The mean frequency of use (number of days spent over one year in forests by activity) in
each municipality is shown in the Table 4.6. This table shows the municipalities surveyed
as grouped by Carvalho Ribeiro and Lovett (2009) with the shading representing the
groups from deep rural (no colour) to inner urban (dark grey). The first column of the table
also shows the classification of the municipalities by Pinto-Correia and Breman (2008).
Despite such a range of uses (recreation, timber and non-timber collections, hunting and
grazing) occurring simultaneously the results of the study show that in some municipalities
the mean frequency of use is overall low. For example, people from Braga (a large city)
spent in 2006 approximately 10 days doing recreation activities in forests, 2 and 1 days
collecting timber products and non timber products, respectively. Uses such as hunting and
grazing are marginal in Braga.

Table 4.6. Mean frequency of use of forests (number of days spent in a year)

Pinto-Correia Municipality N Recreation Timber Non Hunting Grazing

and Breman timber
(2008)
3a Arcos Valdevez 28 11.96 12.29 3.00 15.14 95.89
3a Terras Bouro 32 26.86 1.82 3.95 18.32 82.95
2a Pte Barca 38 5.22 3.72 4.11 3.72 91.58
3b Montalegre 22 3.22 5.74 7.04 11.96 85.0
3a Vieira Minho 23 5.61 13.87 10.04 28.87 115.43
3b Melgaco 9 3.67 17.11 18.22 12.67 163.89
2b Vila Verde 26 4.93 8.90 2.17 2.23 0.0
2b Amares 20 5.67 0.48 1.44 2.04 0.0
2b Povoa Lanhoso 20 8.13 2.35 0.48 5.39 0.0
2b Pte Lima 23 3.81 5.76 15.05 0.38 18.05
1 Barcelos 29 10.28 42.86 26.45 1.24 7.24
1 Esposende 28 23.07 3.89 7.64 .07 0.0
2b Viana Castelo 32 5.69 3.25 1.56 1.25 22.81
3c Braga 29 9.83 2.31 1.41 .17 0.0
Total 205 10.06 13.08 9.26 8.44 72.17
landowners
Total non land 170 8.58 2.64 3.10 3.82 2.88
owners
PCR&B:Pinto Correia and Breman (2008): 1 Production- Specialized agriculture with high profitability
2aExtensive agriculture environmental quality homogeneous landscapes, 2bExtensive agriculture
environmental quality diversified landscapes, 3a Agriculture environmental quality in forested areas 3b
Agriculture environmental quality in mountainous areas, 3cAgriculture environmental quality in urban areas.
Note the shadow represents the gradient from deep rural (no colour) to inner urban (dark shadow)

Figure 4.5 shows the geographic variation in the key uses of timber, recreation and
grazing. As can be seen in both Table 4.6 and Figure 4.5, people in different municipalities
uses forests in order to obtain different goods or services. Municipalities such as
Montalegre, and Arcos de Valdevez (classified as deep rural in Carvalho-Ribeiro and
Lovett 2009) have the highest mean of use for grazing activity, whereas municipalities
such as Esposende and Braga, outer and inner urban, respectively have highest means for

114
recreation activities. This result supports the existence, in terms of forest uses, of an urban-
rural gradient which was identified in Chapter 3. However, this rural-urban gradient, as far
as use of forest is concerned, is more obvious for the Cávado (including municipalities
such as Braga) than the Lima river basin which has high mean frequency of use for grazing
across the area. After highlighting the different character of the two watersheds, it can be
said that more traditional uses such as grazing are predominant in rural areas at the top of
the catchments. By contrast, in more developed areas (except Viana do Castelo and
Barcelos) people use mainly the forest areas as “play grounds” for several recreation
activities.

LIMA
CÁVADO

Figure 4.5. Frequency of use of forests across the study area

Analysis of the frequency of use for each respondent showed that there were different
types of users. Some respondents use forests daily for a variety of goods/services while
other people use forests mainly for recreation. In order to represent this variety a
classification by type of user was created. Respondents that had values above the mean for
more than one activity (e.g. timber collection and livestock grazing) were classified as
multi-users. Respondents with frequency of use above the mean in only one activity (e.g.
recreation) were classified as single users. When frequencies of uses were below the mean
in all activities participants were classified as occasional users. Finally, participants with
115
less than five visits in a year to a forest were classified as non users. The vast majority of
the respondents were either occasional or single users of which the most representative
were the recreationalists. Only 21% (79 out 375) of the respondents used forests in order to
get a multitude of goods/services throughout the year (multi-users) and even for multi
users the income generated via forestry was said to be marginal. Overall, 87.5 % of the
interviewees, in some way, used forest areas. Figure 4.6 shows the way in which the types
of user were scattered across the urban rural gradient. It can be seen that in deep rural areas
the majority of the interviewees were either multi users or single users while in the
remaining socio-economic categories single and occasional users were more frequent.

16%

14%

12%

10%
Multi user
8%
Single user
6% Occasional user
4% Non user

2%

0%
Deep Rural Developing Outer urban Inner urban
Rural

Figure 4.6. Type of users by socio-economic group

It was clear from the survey that the past role of forests as a complement to agricultural
and animal husbandry activities has been declining steadily due to constraints related to
declines in primary sector activity which some authors attribute to CAP policies (Andresen
and Castelbranco, 1993; Firmino, 1999; Pereira et al., 2005). As a result, at present, the use
of forests as an income source is minimal with the majority of forests in the region in a
near-abandonment situation. Only in rural areas, do forests still have importance in
complementing other activities such as livestock grazing (remnants from the past) but these
uses are not providing income to rural populations and several rural villages are already
completely abandoned (Pinto-Correia and Breman, 2008).

4.3.2.2. Preferences for management strategy

Figure 4.7. shows that all the three functions of forests (production, protection and
recreation) were selected as “first choice” in at least one of the municipalities.
116
Figure 4.7. Preference for management strategy across the area

By comparing Figures 4.5 and 4.7 it can be seen that in municipalities such as Barcelos
higher mean frequencies of use for timber collection were consistent with the choice of
production as a management strategy by the majority of the respondents. The same
happened in Esposende which had higher mean frequencies of use for recreation and, in
the vast majority of cases, recreation as the preferred forest management strategy.

Despite an overall trend of agreement between uses and preference for management
strategy across the municipalities Figure 4.7 does not reveal an obvious association
between the level of socio-economic development and preference for management
strategy. Furthermore, when analysing which management strategy was ranked first, the
results were evenly distributed: 102 respondents selected production, while recreation and
protection were chosen by 141 and 132 respondents respectively. It can be seen in Figure
4.8. however, that protection and recreation were priorities of the majority of people living
in deep rural municipalities. The developing rural group was the one in which the
productive function was more often selected.

117
By analysing the ways in which the three functions were ranked by each respondent it can
be seen that 46% of the respondents (87+86=173 out of 375) placed production last when
ranking criteria (last two rows in Table 4.7). This means that a considerable percentage of
people were more interested in recreation and protection rather in productive functions of
forests.

Table 4.7. Rankings of management strategies, its association with criteria for SFM

Code 1st 2nd 3rd N To HV Ti CS SW Bio

PrRcPt Production Recreation Protection 36 4 13 15 1 1 2

PrPtRc Production Protection Recreation 65 4 10 36 6 6 2

RcPrPt Recreation Production Protection 54 7 36 1 8 2

PtPrRc Protection Production Recreation 44 7 25 6 4 1 1

RcPtPr Recreation Protection Production 86 34 17 9 4 14 8

PtRcPr Protection Recreation Production 87 23 37 6 8 8 4

Pr:Production, Rc:Recreation, Pt: Protection, N. number of responses. To:Tourism, HV: Heath and vitality,
Ti:Timber, CS: Carbon sequestration, SW: Soil and water protection, Bio: Biodiversity

Amongst these who ranked production last were the majority of people living in deep rural
areas (Figure 4.8) while productive functions were more important for people living both
in developing rural and outer urban areas.

50
45
40
35
30 Deep rural
25 Developing rural
20 Outer urban
15 Inner urban
10
5
0
PrRcPt PrPtRc RcPrPt PtPrRc RcPtPr PtRcPr

Pr:Production, Rc:Recreation, Pt: Protection

Figure 4.8. Preference for management strategies by socio-economic group

118
The way in which different type of users (multi, single occasional and non-users)
prioritized the management options is shown in Figure 4.9. Putting more emphasis on
protective issues (PtRcPr) were single users (in majority recreationalists).

45

40

35

30

25 multiuser
single user
20
occasional user
15 non user
10

0
PrRcPt PrPtRc RcPrPt PtPrRc RcPtPr PtRcPr
Pr:Production, Rc:Recreation, Pt: Protection

Figure 4.9. Preference for management strategies by type of user group

Table 4.8 shows the preferences for forests characteristics according to the management
strategy chosen. Independently of the management option, a higher percentage of
respondents preferred a mixed rather than pure stands. The preferred type of mixture was
often one of pine and oak though in the case of productive functions eucalyptus*oak was
more common. Uneven stands were clearly preferred over even stands (again, the
difference was less evident in the case of production). Autochthones tree species were
preferred over exotic tree species. The only characteristics that differed depending upon
management option were shape and area of the plots. Respondents that selected recreation
were in favour of smaller plots of irregular shapes. On the contrary, the majority of
respondents that selected production preferred bigger plots.

119
Table 4.8. Forest characteristics for different management strategies

Management Production Protection Recreation

N=375 101 132 140

Stand structure Mixture 72 121 135

Pure 29 11 5

Type of Mixture Pine*oak 48 112 114

Pine*eucalyptus 43 9 23

Stand_age Even aged 34 14 22

Uneven aged 67 118 117

Area_plots Big (>10 hectares) 74 75 44

Small(<10 ha) 27 57 95

Shape_plots Regular 43 35 23

Irregular 57 96 116

Species origin Local 60 129 130

Exotic 38 2 5

Tree species Pine tree 35 18 26

Oak tree 19 107 93

Eucalyptus 47 7 21

4.3.2.3 Trends of change in forest characteristics

At the present time, direct goods/services such as recreation, timber collection and tourism
activities were classified as very important. Regarding indirect goods/services provided by
forests, both air quality and soil and water protection got higher percentages in the very
important category. In 20 years time, the indirect goods/services kept the same level of
importance. The same trend is shown by direct goods such as recreation, timber collection
and timber activity. In the future new goods/ services are expected to be provided by forest
areas in this region. One example is that of a forest contribution to more “green” life styles
by both sequestering carbon as well as providing more “green” sources of energy
(renewable energy). The role of forests in contributing to a more decarbonised life style
increased, in the very important category, from 15.7% at the present to 56% in the future.
Figure 4.10 shows the importance attributed to goods/services at present and in the future
in the study area.
120
 90

80

70

60

50

40 At present
30 In 20 years time

20

10

0
Recreation Collection Collection Hunting Grazing Tourism Air quality Soil and Renewable
timber non- timber (fishing) areas water energy
products products protection

Figure 4.10. Very important goods and services provided by forests (present and future)

Regarding trends in forest condition, according to respondents‟ opinions, degradation of

forest conditions has been occurring in all surveyed municipalities. 80% of the
interviewees (301 out 375 interviewees) stated that in the last 20 years the condition of
forests had become worse. As far as reasons for this are concerned, the results indicate that
wild fires are by far the most important factor (Figure 4.11).

180
160
140
120
100
80
60
40
20
0
Fragmentation Decrease in area Burnt area Simplification of Invasion of exotic
stand structure tree species

Figure 4.11. Threats to SFM in the region

4.3.3. Comparing the field and questionnaire surveys

From comparing the results of the questionnaire with those of the field survey it is apparent
that there is a mismatch between what the public considers as “good forests” and the
forests that exist in the area. The public clearly prefers uneven stands, preferably of oak
121
and pine in which old trees are present. In the area there are some uneven stands however,
these are mostly of pine and eucalyptus and there are few old trees (with diameter above
47.5 cm). The field survey showed that there are still good natural conditions assuring
prosperity of forests in the study area (as indicated by the abundant presence of natural
regeneration). However, enforcing management is likely to be important to overcome some
threats such disappearance of old growth forests and the recurrent presence of litter
(Gibbons et al., 2008).

Across the study area as a whole the number of respondents preferring different
management strategies was roughly similar but priorities varied between municipalities
(Figure 4.7). From this it can be concluded that multifunctional forest management is
needed across the area of the two watersheds. Although there are still multifunctional
forests in the area (e.g. the agro-silvo-pastoral system) (Firmino, 1999), these are remnants
of “old traditional management practices” that are known not to be sustainable at present.
In these areas population is declining, and farming related activities are decreasing (Pereira
et al., 2005; Pinto-Correia and Breman, 2008).

Consequently, the role that forests had in the past as a complement of agricultural and
animal husbandry is not viable at present. Currently, the use of forests as an income source
is minimal in the region with many forests being in a near-abandonment situation (e.g.
litter in the majority of the plots). There is thus a need to forge new or reinvent “old
practices” in order to vary forest roles across the area. In other words, the comparison
between the present condition of forests, public preferences for forests and the provision of
services by forests across the rural-urban gradient highlights the need to move towards
other type of forestry management that can shift from “vicious” to “virtuous” circles
(Matthews and Selman, 2006). According the vast majority of the interviewees (80%) the
condition of the forests in the region has become worse and this clearly calls for action to
reverse such unfavourable trends of change.

As already discussed in Chapter 3 coastal areas had changes in forest area during 1995-
2005, in the range of -10 to +10%, while for inland areas (adjoining Spanish border) the
changes were in the order of -30 to -50%. This increase in area of forests in more
developed municipalities was linked with productive functions (e.g. plantations of
maritime pine and eucalyptus by timber and pulp industries) indicating that dynamics were
already in place to further enhance the rural/urban disparities.

It is evident from this analysis that in order to move away from “a vicious towards virtuous
circle” there is a need to go beyond a “zoning productive functions approach” across the

122
area. Instead, the results indicate that an integration of functions is crucial, especially in
rural areas. For urban areas a spatial separation of functions such as production and
recreation (already in place in urban areas) may be more suitable.

It follows from the above that in rural areas it is crucial to reinforce the strengths of an
integrated type of multifunctionality in order to provide a multitude of goods and services
(timber and non timber products, grazing areas in forests). However, municipalities on the
sea coast are dependent on services such as flood prevention and soil retention from the
forests located in the top of the catchment. This indicates the need to create partnerships
between the areas downstream and the forest stakeholders upstream in order to work
together for whole catchment management.

In order to promote such whole catchment management the research suggests that in the
municipalities located at the top of the catchment there is need to integrate a multitude of
forest functions (production, protection and recreation) while spatially separating the
functions of production, recreation and protection (spatial multifunctionality) or even using
forests as mainly productive in some areas (through dominant use for timber production) is
likely to be more appropriate in the municipalities at the bottom of the catchment. Figure
4.12 shows the ways that different types of multifunctionality could vary across the area.

Figure 4.12. Different types of multifunctionality across the area

123
This distinction can be further interpreted within the framework proposed by Elands and
Wiersun (2001) regarding the roles of forests in Europe. Based on this framework it can be
said that within the study area (Lima and Cávado watersheds) forests in more urbanised
areas are likely to have either utilitarian, hedonist or agri-ruralist roles while in rural areas
their role has more to do with sustaining living communities as well as preserving nature
(community stability and nature conservation). Consequently, while the “basins of
attraction” (Matthews and Selman, 2006) in rural areas are likely to be more related to
conservation and maintaining living communities which may be easily associated with
tourism (which will require a more multifunctional forest) those in urban areas are likely to
be more related to conciliating profitable timber exploitation and recreation (Matthews and
Selman, 2006). It is important, however, to recognise that the traditional way of sustaining
living communities is, at the present, obsolete therefore new strategies ought to be
developed.

Based on the findings of Elands and Wiersun (2001), Slee et al. (2004) and de Blust and
Omen (2000) related to rural development discourses, goods and services provided by
forests and the three types of multifunctionality, respectively, Table 4.9 presents
suggestions regarding forest roles and strategies in different parts of the Minho region.

124
Table 4.9. Role of forests in sustaining landscapes in the region

Carvalho Goods and services Role of forestry Type of Multifunctionality

Ribeiro and provided by forests discourse
Lovett (2009)

Deep rural Timber and non- Community stability Integrated

timber products

Social, cultural Community stability Integrated

identity, traditions

(shadow or halo) Community stability, Integrated/Spatial

Tourism enhancer, Nature conservation

Carbon Nature conservation, Integrated/Spatial

sequestration Community stability

Biodiversity Nature conservation, Integrated/Spatial

Community stability

Water and soil Nature conservation, Integrated/Spatial

protection Community stability

Developing (direct) mostly Utilitarian Dominant use (production)

rural, urban timber
(Inner and
Outer)
Forest biomass for Utilitarian Dominant use (production)
renewable energy

Complement of Agri ruralist Spatial (protection and

agriculture recreation)

(Shadow or halo) Hedonist Dominant use ( recreation)

recreational
Spatial (protection and
activities
recreation)

Related to this the extent to which forest management targets should vary can be further
explored. This suggestion of management targets is based in the assumption that enhancing
protective functions such as water and soil protection as well as promoting tourism in rural
areas will require a faster increase in the patch size of broadleaved tree species. On the
other hand, in urban areas although broadleaves are important for recreational activities the
increase of patch size of broadleaves is likely to be done at slower pace. Furthermore,
timber exploitation in the area is based mainly on maritime pine timber which is either
transformed in local sawmills or it is exported. The other tree species that is becoming
increasingly important in urban areas (inner and outer as well as in developing rural) is
Eucalyptus globulus used mainly for pulp production by the paper industry. Consequently,
because the dynamics in urban areas are towards more productive functions the increase in
patch size of broadleaves will be lower than what would be ideal for rural areas.

125
This recommendation refines what was suggested in the discussion of Chapter 3, namely a
general increase of patch size for broadleaves tree species across the area. In Figure 4.13
the minimum patch size of 5 ha is based on the results of Chapter 3 as this was the mean
value across the area. Based on Forman (1995) achieving a patch size of 30 ha can be
suggested as the target management goal in deep rural areas but this is only presented as
indicative. More detailed ecological and economic valuation needs to explicitly address the
patch size for broadleaves in the area for the fulfilment of the purposed role.

35
30
25
20
15
10
5
0
2000 2020 2040 2060
integrated multifunctionality spatial multifunctionality

Figure 4.13. Management targets suggested for the patch size of broadleaves

4. 4. Discussion

Based on the results presented in this chapter there is a need to vary the management of
forest functions spatially across the area of the two watersheds in order to promote whole
catchment management. Varying the type of functions that forests provide suggests that in
some areas integrated multifunctionality may be the best management option to pursue
(within single forest plots). Whereas in other areas, the landscape may be better suited to
spatial multifunctionality where each plot is managed for one major function (e.g.
production). Nonetheless when the totality of the catchment is considered a multifunctional
landscape mosaic would be present (de Blust and Olmen, 2000).

Although there are still integrated multifunctional forests in the areas located at the top of
the catchment (e.g. the agro-silvo-pastoral system), these are remnants of “old traditional
management practices” that are no longer sustainable. As a result, income from these
forests is minimal and they are in many cases nearing abandonment (Firmino, 1999;

126
Pereira et al., 2005). Thus there is a need to forge new practices or reinvent “old practices”
in order to re-link rural people and forests in a sustainable virtuous cycle (Matthews and
Selman, 2006).

In urban areas, primary and tertiary sectors compete fiercely over land use and regardless
of the type of economic activity the primary concern is profitability and competitiveness
(Niza and Ferrao, 2006). This highly productive trend is already occurring in the more
developed areas in the form of very profitable intensive agricultural systems (Pinto-
Correia and Breman, 2008) and the exploitation of forests for timber and pulp production.
However, these highly production orientated systems which are occurring in the
municipalities located near to the coast are dependent on ecological services such as flood
prevention, soil retention and water flow maintenance provided by forests upstream.

It follows then, that there are clear opportunities for good implementation of the WFD in
the two catchments studied. The WFD clearly calls for partnerships that enable whole
catchment management, this would require the establishment of partnerships that span
different government institutions (e.g. municipalities) and should include the private
sector to overcome some of the problems highlighted in rural areas whilst also addressing
the needs of urban communities.

In addition to the WFD, a range other of measures such as: ownership or management of
the land via long term leases; regulatory controls; monetary disincentives to discourage
undesirable uses; financial incentives to encourage desirable uses and other voluntary
methods (e.g. advice, demonstration) could be combined in order to influence “desirable”
action on the ground (Gilg, 1996 cited by Selman 2006: 128). Thus a set of legally binding
mechanisms combined with an appropriate set of measures (for example the “menu”
presented above) could be mixed to create win-win partnerships so that urban areas may
gain by investing in forestry activities located in upper reaches of the catchment thereby
reducing flood risk, water shortages and water purification costs. These types of
arrangements are already in place elsewhere in Europe (Group Perrier Vitel) and the USA
(e.g. New York City).

However, in order to setup a useful framework for whole catchment management and the
establishment of effective partnerships there will be a need for an appropriate mix of
incentivisation, regulation and extension work (Selman, 2006). Incentivisation is
associated with governmental support to private operators (e.g. farmers) to practise
desirable land use practices in the forms of subsidies or grants. Regulation on the other
hand controls by enforced penalties for non compliance. Finally, extension work seeks to

127
improve and actively promote information, demonstration and extension services in the
belief that this can lead to improved practice (Selman, 2006).

In addition to the above, as can be seen in Table 4.13, in order to implement these different
roles for forests there will need to be an integration of forestry with at least three other
sectors namely tourism, agriculture and energy (forestry biomass, carbon sequestration).
Forests have to be regarded as one component of the whole landscape mosaic. This raises
the need to integrate forestry into formal planning mechanisms that merge different
sectoral interests. In order to efficiently tackle the trends of depopulation and population
ageing forestry needs to be more competitive and fully embrace the ecosystem services it
provides.

It is true that there is a need to deal with incipient “markets” in which uncertainty is
enormous (Dwyer, 2007). Furthermore, the OECD (2006) has already pointed out that the
ways to move forward should be based more on investments and partnerships across places
instead of relying in subsidies and grants as occurred in the past. In such a diversified
Europe (Pinto-Correia and Breman, 2008) there is a need to embrace the challenge of
creating, maintaining and enforcing new types of partnerships to face uncertain futures. For
this to happen, the WFD and its river basin plans could have a crucial role. So far, there are
not institutions in place that can lead the implementation of whole catchment management
but the WFD is likely to help in this.

This research also suggests that it is of upmost importance to tackle the problem of the
wild forest fires which are a matter of concern within the whole region and need to be
addressed both in rural and urban areas. The severe wild fires in the mountains of
neighbour Galicia in 2005 have already demonstrated that fires upstream may undermine a
million Euro sea food business located downstream
(http://www.euroresidentes.com/Blogs/2005_08_01_archive.html) [accessed online on 14
March 2009]. This ecological disaster was widely discussed in the Portuguese media and
there is public awareness of the need to tackle forest fires at the river basin scale. In
addition to the public awareness of the need to avoid massive forest fires, sustainable
forestry management is also a stated priority for the Portuguese Government (DGRF,
2007a, b; INAG, 2005, 2006). These two premises seem to be a good basis to move
towards sustainable forestry management at the catchment scale.

However, the implementation of the WFD and the participation/governance mechanisms

inherent to it are far from being easy to achieve (Bateman et al., 2006). There are a number
of pitfalls for participation, including the practical challenge of involving multiple

128
viewpoints without focussing too much on individual and personal biases (Milligan et al.,
2009). In practice, balancing the interests of multiple stakeholders is extremely difficult
(Adger et al., 2004). The Water Framework Directive will require governance systems that
deal with very sensitive issues such as land ownership because the vast majority of the land
within the study area is privately owned (DGRF, 2007a). Therefore, mechanisms able to
manage the whole landscape (Dolman et al., 2001) transcending property boundaries need
to be agreed upon by land owners and governments. These may be addressed by separating
ownership of the land from management via long term leases, but conditions need to be
agreed upon by all the stakeholders involved and economic incentives from governments
are likely to be required at least in the initial stage of the process (O' Riordan and Stoll-
Kleemann, 2002).

As previously stated, the river basin plans to be put into place by the end of 2009, clearly
need to negotiate a set of commonly agreed objectives as well as a framework of shared
responsibilities (CEMAT, 2007; INAG, 2005, 2006). This certainly requires the
establishment of public-private partnerships in order to be successful (Dwyer, 2007). It will
also call for a more effective integration of forestry with other activities such as tourism or
even with sectors such as energy (Dwyer, 2007). However, defining the specific role that
forests may have in different places within the area of river basins seems to require a more
fine scale of analysis. In other words, because within the area of a single municipality the
roles of forests greatly vary, it appears important that further research needs to focus on
smaller scales for example at the parish level.

4.5. Conclusion

It is likely that forests will have different roles in contributing to sustainable landscapes. In
rural areas, forests are needed to provide a multitude of goods and services that supply
direct goods for rural communities and assure environmental protection for downstream
areas. The results of this study indicate that different “spatial” approaches for forestry will
be required in different locations within and between watersheds in order to strengthen
“virtuous circles” across an urban-rural gradient in Northern Portugal. It has been argued
here that achieving sustainability in urbanized landscapes will require different strategies
from in rural landscapes (Antrop, 2004, 2005, 2006). However, it is also likely that the
sustainability of rural and urban landscapes will depend on each other, and strategies that
strengthen rural/urban patterns of development will be important (Johnson et al., 2002).
The research also suggests that the implementation of multifunctionality in forestry greatly
depends on the integration of forests within the whole landscape mosaic (Dolman et al.,

129
2001). It is also apparent that it is crucial to create and reinforce partnerships amongst
forest stakeholders downstream and upstream of each other. In this context, European
policies such as the WFD may provide the means to move towards sustainable forestry
management in the Minho Region of Northern Portugal.

130
CHAPTER 5. PUBLIC OPINION REGARDING
ATTRACTIVENESS AND MANAGEMENT OF FOREST
LANDSCAPES: PREFERENCES FOR FOREST COVER AND
STAND STRUCTURE

131
Abstract

The role of forestry is changing from an emphasis on production to also include

consumptive (e.g. recreational activities in aesthetically pleasing forest areas) and
protective (e.g. providing biodiversity and avoiding floods and soil erosion) functions. Due
to the ability of forests to provide an array of these services their role has been increasingly
recognised as being of utmost importance in sustaining landscapes. There is an ongoing
debate about the ways in which aesthetic preferences may threaten the ecology of natural
ecosystems, but, the effects of stakeholders‟ preferences of “good management” on the
ecological functioning of forests has not received much attention. This study examined
public aesthetic and management preferences for forest cover and stand age. Both verbal
and visual approaches were used. By calculating the degree of correlation between two
rankings of a set of photos showing i) different area of forest cover and ii) stand ages based
on attractiveness and management criteria, the study found that the vast majority of the
public ordered the two sets of photos similarly independent of the criteria under
consideration. This suggests that the majority of the public consider a “beautiful forest”
also a “well managed forest”. However, differences in both the rankings of attractiveness
and management were found across user groups. In addition, there were contrasts in
preferences for management strategies that could compromise the ecological functioning of
forest ecosystems. This suggests that preferences for forests from both public in general
and specific stakeholders need to be carefully addressed in planning and more work needs
to be done to achieve a whole catchment approach. Above all, planners, researchers,
decision makers and the general public need to work together in order to fully embrace the
challenge of multifunctional forestry management. In the research it was also apparent that
the use of the photos was critical; if only verbal data were used there were no conclusive
findings about public preferences for forest cover.

Key words: attractiveness, management, multifunctional forests, rankings, photos,

correlation

132
5.1. Introduction

Previous research has established that there is a difference between human preferences for
natural resources and the “ecological status” of those resources (Gobster et al., 2007;
Sheppard and Harshaw, 2000). The disjuncture between human preferences and the
“ecological status” of the environment was termed by Gobster et al. (2007) as the
“aesthetics-ecology debate”. The authors argued that visually appealing landscapes are not
always ecologically healthy, in the same way, “ecologically healthy landscapes may not be
aesthetically pleasing” (Gobster et al 2007: 962). A similar argument was previously
developed by Sheppard and Harshaw (2000). In their work some correspondence between
aesthetics and ecology was found but was not universally present “it can be seen that what
is believed to be ecologically good may not look good, and that what looks good may not
be ecologically sustainable” (Sheppard et al. 2000: 284). Selman (2006) also raises the
issue by addressing what he called “nature-society” debate. According to Selman one
practical problem for landscape planning is whether there are fundamental differences or
similarities between the environmental requirements of wild species and the landscape
desire of humans. He argues that is likely that people prefer a) tidiness and b) tend to avoid
“landscapes of fear” whereas nature a) often likes “scruffiness” and b) requires natural
disturbance such as fire and landslides (Selman, 2006: 63).

This debate is of major importance in order to plan (establishing goals and policies) and
manage (putting planning goals into practice) towards sustainable environments (Gobster
et al., 2007). Acknowledging such importance, the study of aesthetic preferences or scenic
beauty (Daniel, 2001) in relation to the characteristics of natural environment is a recurrent
topic in the literature (Thorne and Huang, 1991). Despite a vast repertoire of studies
reporting public aesthetic preferences for agrarian systems (Rogge et al., 2007b),
hedgerows and other natural features (Evernden, 1988; Misgav, 2000) forests are a
recurrent topic in the literature (Ribe, 1982, 1989).

As far as forests are concerned, in addition to aesthetics and ecology issues, there is also a
third dimension which is related to the ways in which the productive functions of forests
are achieved. Associated with productive functions of forests, in addition to timber, there
are a variety of other non-timber goods such as honey and game that are likely to be
important for the livelihoods of the communities that live around the forests. The stand
composition and structure as well as the arrangement of forest patches can enhance or
detract from the ability of the forest ecosystem to provide such goods (Kellomaki and
Pukkala, 1989). For example, a forest in which ground vegetation is removed in order to

133
reduce competition between species (trees and understorey) for increasing timber yield will
lack appropriate refuge for game; furthermore, because ground (and floral) vegetation will
be minimal, honey production will be marginal. Ground vegetation removal is one example
of forest management practices that affect the provision of both timber and non-timber
goods. Recognising the importance of forest management in the delivery of ecosystem
goods and services, there also studies exploring public preferences for practices such as
thinning (Silvennoinen et al., 2001), afforestation styles (Karjalainen and Komulainen,
1998) and clear cutting types (Rekola and Pouta, 2005) in order to improve management
programs (Kellomaki and Pukkala, 1989).

It follows from the above that, in the forestry arena, preferences for forest characteristics
that can enhance the provision of certain goods or services are important in order to
manage towards sustainability. Public preference for forests is defined by Sheppard and
Harshaw (2005:7) as “the degree to which a person or group prefers a situation or feature
over other situations or features”. These authors argue that scenic beauty (aesthetics) and
preferences for management strategy (management) can be distinct. For example, a
livestock grazer may find a dense and contiguous forest aesthetically pleasing, while
acknowledging that his/her own livestock will not be able to thrive in such habitat, thus
making it likely that he or she will not favour continuous cover forestry as a management
strategy, at least in the area he/she grazes livestock (Ribe, 2002). Consequently, one of the
reasons for the complexity of addressing public preferences is the difficulties in
disaggregating “the eye from the beholder” (Sheppard et al., 2000).

Although huge efforts have been made to explore attractiveness of forests as well as public
opinion about forestry management practices per se, there are few cases in which the
differences between attractiveness and management preferences for forests have been
explored in the literature. In the majority of the work done so far, research methods
conclude that public aesthetic and management preferences are positively associated. For
example, a recent study by Surova and Pinto Correia (2008) reports divergent preferences
for cork oak montado characteristics among hunters, mushroom pickers, beekeepers and
landowners in Southern Portugal. For example, hunters preferred forests with ground
vegetation because they provide “good quality” hunting reserves while beekeepers
preferred forests where floral composition was richer because it was associated with
“better” quality honey (Surova and Pinto-Correia, 2008).

A review of public preferences for forests indicates that preferences are likely to vary with
personality and socio-economic factors such as gender and landownership (Abello and
Bernaldez, 1986; Tips and Vasdisara, 1986), professional background (Rogge et al., 2007b;
134
Winter, 2005) and type of recreational activity (Harshaw et al., 2006; Roovers et al., 2002).
In the same way it was found that farmers, experts and country dwellers have different
perception of agrarian landscapes in Flandres (Rogge et al., 2007b). Another study
reported differences and conflicts between preferences of “local and extra-local”
inhabitants of protected areas (Zube, 1986).

As well as examining preferences for forests across stakeholders groups, this study also
investigated the extent to which public aesthetic and preferences for management strategy
are related. Previous research on this topic has generated varied findings. Tahvanainen et al
(2001) found that scenic beauty and recreational preferences differed considerably from
each other. Another study found that recreation ratings were related to, but different from,
scenic ratings (Shelby et al., 2005). Differences between aesthetic preferences and
acceptability of management options were also found by Ribe (2002).

In order to plan for sustainable landscapes where forests are an integral part, some
combination of the functions of protection (ecology), recreation (related to aesthetics) and
production (related to management practices) is required. How then can planning and
management integrate scenic beauty, ecological and management goals together? With
increasing recreation pressures occurring on woodlands and forests should planners
“favour” special management types? If so, what are the ecological implications?
Furthermore, preferences for what types of activities should be represented in forest land
use planning? (Harshaw et al., 2006)

It can be argued that rather than accounting for all needs separately, planning and
management towards sustainable environments needs to focus on ways in which ecology,
aesthetics and management are consistent with each other. Some studies have identified
compatibilities between forest uses, e.g. timber production and water quality , but it is also
likely that trade-offs between wood production and other forest products and values may
need to occur (Stevens and Montegomerey, 2002). In this situation, there is a requirement
to engage with the public in order to create awareness that their preferences for
environmental characteristics might interfere with ecosystems and other functions. As a
consequence, approaches to plan for sustainability need to be able to communicate
explicitly to the general public in order to provide evidence that in order to provide
“invisible” services such as soil and water protection ecosystems may have to be less
attractive and/or provide less direct goods such as timber or other non-timber products.

This challenge is one of the reasons why the distinction between aesthetic and management
preferences is very important when trying to achieve multifunctional landscapes that can

135
support sustainability goals (Mander et al., 2007). Active involvement of “lay” people and
their organizations in planning for sustainability is desirable in many ways (Selman, 2006).
A key principle underpinning this study was therefore that members of the public and
stakeholders need to be aware of the implications that their preferences (for both aesthetics
and management) might have on the ecology of natural resources.

Two widely reported threats to forest in temperate regions are fragmentation of forest
landscapes and simplification of stand structures (Jongman, 2002). Consequently, in order
to attain public participation and engagement in the design of sustainable landscapes public
preferences for these two parameters of forests are crucial (Sheppard, 2005; Sheppard and
Harshaw, 2000, 2001). This study specifically investigated people‟s preferences for
attractiveness and management for forest as far as percentage of forest cover and stand
evenness were concerned. The former was used as an indicator of fragmentation and the
latter stand simplification (Fahrig, 2003). These variables were investigated in the setting
of northern Portugal where forests are an important landscape feature whose management
has faced a number of challenges (Firmino, 1999; Pinto-Correia, 2000). One of the key
issues for forest management in Portugal, as well as in other Mediterranean countries, is to
reduce the impact of wild forest fires. If it is known that fragmentation is a major threat to
forests, it is also acknowledged that contiguity of forest stands causes problems in fire
combat, and consequently discontinuities have been created in order to reduce the risks of
fire spread. Another measure to reduce fire impacts is to create variations in stand
structures since the different wood densities delay spread more easily than homogeneous
stands. These factors need to be taken into account when interpreting attitudes to variables
such as forest cover and stand ages in a country such as Portugal.

Public preferences can either be studied throughout verbal and visual approaches
(Tahvanainen et al, 2001). Because technical concepts such as forest cover and stand
structure can be unfamiliar to respondents the questions were addressed both verbally and
visually. The former was used because when asked first verbally the respondents had the
opportunity to “frame” their own view and meaning of the concept and this may
contribute, consciously or not, to a better “framing” of what he/she has been asked to
answer. The visual approach is, per se, one of the most used techniques due to its ability to
set the frame and put everyone‟s mind in the same context (Tahvanainen et al., 2001). Box
5.1 shows the main questions addressed by the research.

136
Box 5.1. Questions
1. Do public preferences for level of forest cover and stand structure vary
according to whether attractiveness or management objectives are considered?
2. Are there differences in aesthetic and management preferences across different
user groups?
3. Are there indications that public preferences might threaten the ecological
functioning of forests?
4. Are the results similar independent of whether verbal or visual approaches were
used?

5.2. Material and methods

5.2.1. Study area

River basins are claimed to be appropriate units to study forest ecosystems due to their
ability to link together cultural and biophysical systems and thus represent a suitable scale
to address landscape issues (Selman, 2006). In Northwest Portugal, within the Lima and
Cavado watersheds an urban-rural gradient has been identified by Carvalho Ribeiro and
Lovett (2009). This research indicates that the socio-economic characteristics of the
municipalities in Lima and Cávado watersheds ranges from deep rural in inland
mountainous areas to urban areas (outer and inner urban) located mainly in the coast.

In order to distinguish public preferences for level of fragmentation and stand structure
across the urban-rural gradient a questionnaire survey was conducted in all fourteen
municipalities within the area of these two watersheds. The Lima and Cávado river basins
are located in the Minho region of Northwest Portugal and flow from a mountainous region
in which the highest peak is 1530 meters high to the sea coast along a distance of
approximately 80 kilometres. The mountainous regions are predominantly rural whereas
areas near the sea coast are clearly urban. The study area is mainly under Atlantic climatic
influence, though some areas show Mediterranean microclimate. Broad-leaved trees such
as oak (Quercus robur) are characteristic of the zone, but the most predominant tree
species is maritime pine (Pinus pinaster) while in recent years the area of eucalyptus trees
(Eucalyptus globulus) has been increasing steadily. The majority of the land is privately
owned although some communal lands are found in the mountainous areas.

5.2.2. Questionnaire survey

Public preferences were studied through a questionnaire survey which investigated

opinions regarding i) “attractiveness” and ii) “management” of forests as reflected in
variations of area covered by forests and stand structure. The survey also collected details
137
of the individual frequency of forest use in activities such as timber and non-timber
product collection, recreation, grazing and hunting. During the planning phases of the
questionnaire a meeting with the forestry office of each municipality was arranged in order
to identify contacts for people in the forestry sector. These forestry stakeholders were then
included in the sample. When these contacts were made a “snow ball” approach was used
in order to enlarge the sample. In addition, a street survey was carried out in the town
centre of each municipality in order to collect information from non-users. The initial part
the questionnaire focussed on verbally expressed preferences for forest characteristics (see
Card D in Annexe 3). This was followed by the respondents ranking sets of photographs
according to attractiveness and management criteria (see visual approach-photos shown in
Annexe 4). Because the concepts of attractiveness and management can have different
meanings for different people respondents were told in the attractiveness case “please rank
this set of photos according to the visual appeal of the forests shown”. For the management
criteria respondents were asked to rank photos “according to the characteristics of forests
that most favour your lifestyle and are considered by you to be good management”.

In the initial part of the survey respondents were asked verbally to select one from two
contrasting characteristics i.e. i) preferences for continuous forest cover or a patchy forest
ii) preference for even or uneven stands. This question was repeated for attractiveness and
management criteria. In the latter they were asked to rank five photos showing areas with
different amounts of forest cover and five photos showing contrasting stand structures.
Each respondent ranked from 1 to 5 the photos for fragmentation (labelled FA to FE) and
those for stand structure (labelled SA to SE). This task was carried out twice, once
evaluating attractiveness and secondly management. In the first case 1 represented the
most attractive and 5 the least attractive. In the second 1 represent the best management
and 5 the worst. In both sets of photos there was one image (FE and SE) where controlled
fire had been used as a forest management tool. The questionnaire work was conducted
between March-August 2007 and obtained a total sample of 375 responses.

5.2.2.1. The questionnaire images

In order to study people‟s opinions on different fragmentation levels and stand structures
typical scenes representing the forests in the study area were chosen. Photos from the study
area were considered most appropriate to match with the familiarity of respondents. Pine
forest was selected for all the scenes since, despite the presence of oak in rural areas, it is
the most common tree species in the watersheds. In addition, using only pure pine stands
provided a “control” for species diversity. All the photos were taken in September 2005
(thus not varying the season) using a Nikon digital camera with a 28 mm lens. The scenes
138
used for the study of forest cover were distant views of forests in a mountain setting
whereas for stand structure purposes the images were closer up. Both set of pictures were
taken at the eye level of the observer; pictures showing different levels of forest cover were
in landscape orientation while the photos showing stand structures were in portrait.

The digital images were imported into the Adobe Photoshop image-processing software
and image manipulation tools were used to eliminate all the features such as roads, urban
areas and other unwanted features likely to distract respondents from the factors under
study. This photo manipulation was done taking care not to diminish the “realism” of the
scenes. In addition, all the background colours were set to approximately same value in
order to show only variations in the fragmentation and stand structure parameters (e.g.
setting the same light and sky colours). However, it was decided not to set a blue sky in the
photos with even stands because it reduced the realism of the photos. Instead a grey sky
was used. The final digital images were printed in A5 format and laminated in order to be
kept in good condition throughout the survey.

5.2.2.2. Statistical analysis

Data from the questionnaire survey were analysed in SPSS. Based on frequency of use of
forests (questions 1 to 4; in the questionnaire; see Appendix 3) participants were classified
into groups, e.g. recreationalists vs. non recreationalists, timber and non timber collectors.
Due to the variety of use patterns a further classification was created: respondents that had
frequencies above the mean for more than one activity (e.g. timber collection and livestock
grazing) were classified as multi-users while respondents with above-average frequency of
use in only one activity were classified as single users. When frequency of use was below
the mean in all activities, participants were classified as occasional users. Finally,
participants with fewer than five visits per year to a forest were classified as non-users.

Consistency of the rankings amongst user groups was studied by using the Kruskal Wallis
test. In addition, in order to investigate if there were correlations between the rankings of
attractiveness and management a Spearman rank correlation for each respondent was
calculated. When calculating the correlation coefficients two strategies were followed by
either including or removing the photos (FE and SE) depicting burnt forests. This was done
to assess the sensitivity of the results to the inclusion of such images. In order to compare
the rankings of attractiveness and management, photos showing different levels of forest
cover were subtracted (e.g. value of the ranking for attractiveness FA-ranking of
management FA). The values of the differences were squared. The squared differences for
each photo were summed, multiplied by six and divided by either 120 or 60 depending

139
whether or not FE was considered (number of rankings was either n=4 or n=5), finally, this
value was subtracted from one according to the Spearman rank correlation formula below.
The same procedure was used for the photos showing different stand structures (S photos).

Spearman rank correlation formula. di = difference between attractiveness and

management ranking n=either 4 or 5 depending whether or not burnt forests (FE and SE)
were included.

5.3. Results

5.3.1. Overall results

From the total sample of 375 questionnaires 213 of the interviewees were male (57%) and
162 (43%) were female. The vast majority of the respondents were either occasional users
or single users, of whom the most common were recreationalists. 79 people visited forests
for a multitude of uses throughout the year (multiuser). Overall, 87.5 % of the
interviewees, in some way, used forest areas and 47 (12.5%) did not use forests on a daily
basis (non-users) (Figure 5.1).

Figure 5.1. Number of respondents by type of users

The upper and intermediate areas of Lima and Cávado watersheds are predominantly rural,
and as would be expected, the majority of respondents 68% (257 out of 375) had a rural
background (deep and developing rural categories). The major cities are located near the
sea coast occupying a small area of the two catchments, therefore only 31.4% of
respondents (118 out of 375) had a urban background (outer and inner urban)

140
 Figure 5.2. Number of respondents by area of residence

Figure 5.3 and Table 5.1 show, respectively, the photos used in this survey and the overall
rankings by respondents. The photos are also presented in a larger format in Appendix 4.
The analysis of the mean rank for the set of photos showing different levels of
fragmentation indicated that there was a general trend for more fragmented forests to be
regarded as less attractive. However, it was not the photo showing the most connected
forest (FC) that got best ranking (i.e. lowest score). Instead, the photo showing some
discontinuity of forest cover (FA) was ranked as both the most attractive and best
managed. Therefore, the type of forest that is likely to both fulfil aesthetic and
management requirements in the study area is one that shows some degree of discontinuity
of forest cover (FA).

Photo FC showing a contiguous forest got a better ranking for attractiveness than
management implying that this type of forest, though aesthetically pleasing might not be
considered good management. This is probably due to its association with easier fire
spread.

141
 FC FA FD FB FE

SB SD SA SC SE

Figure 5.3. Photos included in the survey

Table 5.1. Mean rank of attractiveness and management for each photo

FOREST COVER STAND STRUCTURE

(from higher to lower percentage (from uneven to even stands)
of forest cover)

Photos FC FA FD FB FE SB SD SA SC SE

Attractiveness 2.03 1.69 3.00 3.55 4.63 2.09 2.60 2.57 2.91 4.71
Mean rank

Variance 1.411 .611 .693 .865 .977 1.350 .930 1.501 1.534 .841

Management 2.38 1.73 2.88 3.34 4.21 2.28 2.84 2.04 3.38 4.16
Mean rank

Variance 2.032 .970 .964 1.530 2.074 1.229 1.181 1.723 1.450 2.119

In Table 5.1 lower scores represent better rankings and higher ones worse. As far as
attractiveness of stand structures is concerned the photo that got the best score was an
uneven stand structure in which ground cover by shrubs was minimal (SB) while photo SA
showing an even stand with no shrubs at all was rated best for management.

As also shown in Table 5.1, in general higher variances occurred in the ratings of
management than attractiveness (the exception is SC with a higher variance for
142
attractiveness). Therefore, the results suggest that preferences for management strategy are
likely to be more individualistic than assessments of beauty. This presumably occurs either
because people do not fully understand the meaning of management (despite the efforts
made in survey implementation) or because attitudes to management greatly depend on the
way individuals interact with forests (e.g. for recreation or timber collection) and different
users value management in contrasting ways thus increasing the variance value.

5.3.2. Correlation coefficients of attractiveness and management rankings

In order to investigate if there were correlations between the rankings of attractiveness and
management a Spearman rank correlation for each respondent was calculated. Nineteen
respondents did not give valid answers for both questions e.g either not ranking for
management or attractiveness criteria, therefore, the sample size for the analysis of the
spearman rank correlation was 356 individuals.

5.3.2.1. Percentage of forest cover (F photos)

The results in Figure 5.4 show that 43% of respondents (156 out of 356) ordered all five
photos in exactly the same way independently of the criteria under consideration. In this
case the Spearman rank correlation was 1 (rs=1). A strong positive correlation (0.5<rsF<1)
between the rankings of attractiveness and management was found for 73% of the cases
(263 respondents). There was no correlation (rs=0) between the two rankings in 3% of the
cases (11 people) and a negative rank correlation (rs<0) occurred in just 5 % of the cases
(19 out 356). Altogether, the vast majority of the respondents (91%) ordered the rankings
of the five photos FA to FE in the same way independently of the criteria at stake.
When photo FE showing burnt forests was excluded (n=4) the percentage of respondents
with positive correlations decreased to 81% (291 out of 356). There was no correlation
(rs=0) in 1% of cases (5 out 356) and negative correlations occurred with 17% of
respondents (60 out of 356). This suggests that the consistently low rating of photo FE
enhanced the positive correlation between the two sets of ratings.

143
 Photo FE Excluded
All Five Photos

Figure 5.4. Spearman rank correlations for forest cover including and excluding FE

As can be seen in the last column in the Table 5.2, users of forests (multi, single and
occasional users) were more likely to make a distinction between attractiveness and
management criteria. Of the 60 respondents that had negative correlations when rating the
four photos only four were non users. 29 respondents (17+12) lived in rural areas and 31
(16+15) lived in urban areas.

Table 5.2. Type of user and place of residence of respondents with negative correlations in the F
photos ranking

Total % of
Deep Rural Developing Outer Inner Total
category
(135) Rural (104) Urban Urban
fringe (56) Fringe (61)

Multiuser 4 5 4 0 13 17.3%
(75)

Single user 6 2 10 5 23 19.3%

(119)

Occasional 5 5 2 8 20 17.0%
user (117)

Non user 2 0 0 2 4 8.8%

(45)

Total 17 12 16 15 60

Total % of 12.6% 11.5% 28.5% 24.5% 16%

category

144
5.3.2.2. Stand structure (S photos)

Figure 5.5 indicates that 30% of respondents (130 out of 345) rated the five stand structure
photos identically (rs=1) independent of the criteria under consideration. In total, 86% of
respondents had a positive correlation coefficient. A negative spearman rank correlation
occurred in 10 % of the cases (36 out 356). When photo SE was excluded 78% (279) of the
cases still had positive correlation coefficients and 40 % (144 respondents) had a value of
1. In 18% of the cases (65) a negative correlation occurred.

All Five Photos Photo SE Excluded

Figure 5.5. Spearman rank correlations for stand structure including and excluding SE

The respondents that had negative correlations (i.e. differentiated between criteria) in the
assessment of stand structures were mainly recreationalists (single and occasional users) of
which a considerable percentage lived in urban areas. Table 5.3 shows these results.

145
Table 5.3. Type of user and place of residence of respondents with negative correlations in the S
photos ranking
Total % of
Deep Developing Outer Inner Total
category
Rural Rural (104) Urban Urban
(135) fringe (56) Fringe (61)

Multiuser 2 1 1 1 5 6.6%
(75)

Single user 6 4 10 6 26 21.8%

(119)

Occasional 3 11 6 6 26 22.2%
user (117)

Non user 2 3 0 3 8 17.7%

(45)

Total 13 19 17 16 65

Total % of 9.6% 18.2% 30.3% 20.2% 18.2%

category

Summarising, the results presented in this section show: 1) that the photos depicting burnt
areas (FE and SE) were consistently ranked as the least attractive and with poorest
management. As a result, their inclusion in the rankings influences the results by
increasing the positive correlations. The generally strong positive correlation between the
two rankings indicates 2) that there are no differences in preferences for forests according
to attractiveness and management criteria. This finding implies a negative answer to the
first question in Box 5.1, namely that there are not differences in public preferences
depending on the criteria under consideration. Nevertheless, despite the general positive
association there were cases in which negative correlations occurred. As shown in Tables
5.2 and 5.3) people that use the forests regularly (multi and single users) were more likely
to make differences in their assessments. These differences in preferences between user
groups are considered further below.

5.3.3. Attractiveness vs. management across user groups

In the research literature the interrater and interclass correlation are the most used
techniques to assess reliability of repeated measurements such as rating of photos by
different people or at different times (Palmer and Hoffman, 2001). With the aim of
investigating if there were differences in the rankings amongst type of users as well as
according to places where people lived a Kruskal-Wallis test was used. The results of the

146
analysis by type of user are presented in Section 5.3.3.1. Section 5.3.3.2. explores the
results by geographical area.

5.3.3.1. Type of users

Table 5.4 indicates that respondents who regularly used the forests (multiple and single
users) gave lower rankings (higher approval) to the photos showing a discontinuity of
forest cover (FA and FD) for both attractiveness and management. The average ratings for
these two photos were significantly different across the user groups. People who made
little use of the forests (non and occasional users) rated a continuous cover (FC) as more
attractive than regular users but the difference in ranks was not statistically significant.

Table 5.4. Kruskal-Wallis tests for the rankings of attractiveness and management across user
groups (F photos)

Photo FC FA FD FB FE

Mean Att. Man. Att. Man. Att. Man. Att. Man. Att. Man.
rank

Multi user
(79) 2.20 2.59 1.62 1.66 2.76 2.77 3.46 3.42 4.73 4.00

Single
user (129) 2.06 2.33 1.60 1.73 2.98 2.68 3.58 3.17 4.61 4.20

Occasion
al user
(120) 1.90 2.25 1.83 1.85 3.21 3.14 3.51 3.46 4.53 4.31

Non user
(47) 2.00 2.49 1.66 1.53 2.89 2.94 3.68 3.40 4.77 4.34

Sig
Kruskal-
Wallis 0.205 0.341 0.03 0.05 0.01 0.001 0.63 0.46 0.847 0.575

Note: Values in bold are statistically significant at the 0.05 level. Att: Attractiveness, Man:
Management

Table 5.5 presents a similar analysis for the S photos. Only photo SB showing an uneven
stand produced a statistically significant difference in attractiveness ratings, with the more
regular users preferring this image. With the ratings for management there were more
significant differences, one example being the “park-like” forest shown in SA which got
better rankings from people who do not use forests on a daily basis, while multi-users were
least likely to consider such a forest well managed.

The photo showing a burnt forest (SE) got the best assessment from the multi-users
suggesting that people who use forests daily recognize fire as a management strategy in

147
order to reduce the intensity of wild forest fires. There was a statistically significant
difference in the rankings of this photo across the user groups.

Table 5.5. Kruskal-Wallis tests for the rankings of attractiveness and management across user
groups (S photos)

Photo SB SD SA SC SE

Mean rank Att. Man. Att. Man. Att. Man. Att. Man. Att. Man.

Multi user
(79) 2.00 2.33 2.46 2.82 2.80 2.46 2.94 3.17 4.59 3.64

Single
user (129) 1.94 2.20 2.71 2.83 2.48 1.95 2.95 3.46 4.81 4.35

Occasional
user (120) 2.24 2.31 2.67 2.99 2.53 1.92 2.82 3.46 4.67 4.19

Non user
(47) 2.30 2.34 2.40 2.51 2.52 1.91 2.96 3.36 4.77 4.43

Sig.
Kruskal-
Wallis 0.022 0.659 0.058 0.041 0.296 0.007 0.778 0.657 0.319 0.012

Note: Values in bold are statistically significant at the 0.05 level. Att: Attractiveness, Man:
Management

5.3.2.2. Attractiveness vs. management preferences across the rural/urban

gradient

Figure 5.6 shows the variation in the mean rank values for the photos showing different
levels of forest cover across the socio-economic groups of municipalities. There was
variation in preferences across the urban/rural gradient with respondents from urban areas
favouring more connected landscapes e.g. as reflected in the lower mean rank for the
attractiveness criteria for photo FC in the inner urban group. Photo FA was particularly
preferred in the rural areas and in the Deep Rural category the difference in photo rankings
was statistically significant (p=0.03).

148
 5.00

4.50

4.00

3.50
Deep rural
3.00 Developing rural

2.50 Outer urnan

Inner urban
2.00

1.50

1.00
FC FA FD FB FE

Figure 5.6. Public preference for fragmentation level by municipality group in terms of
attractiveness

Photo FA (with an intermediate level of connection) was consistently ranked as the best
with respect to forest management (Figure 5.7).

5.00

4.50

4.00

3.50
Deep rural
3.00 Developing rural

2.50 Outer urnan

Inner urban
2.00

1.50

1.00
FC FA FD FB FE

Figure 5.7. Public preference for fragmentation level by municipality group in terms of
management criteria

Figures 5.8 and 5.9 show the results for stand structure preferences across the urban-rural
gradient. It can be seen that the mean ranks for attractiveness criteria across the rural-urban
gradient are between 2 and 3. An exception is photo SE which got a mean rank of 4.75. A
Kruskal-Wallis test showed statistically significant differences across the socio-economic
149
groups for photos SA, SB, SC and SE. The photo showing an uneven stand (SB) got lower
mean ranks (rated more attractive) by rural rather than urban dwellers. On the contrary
photo SC (showing an even stand) was found more attractive by urban respondents. Figure
5.8 also shows that in rural areas (deep and developing rural) the mean rank values for four
of the photos are more widely distributed whilst in urban areas the mean rank value of the
photos is more similar.

The ratings for management in Figure 5.9 show that in all areas photo SE was considered
to be the least well managed. However, this image got better mean ranks (lower scores) in
rural than in urban areas (p=0.035) implying that rural inhabitants are more likely to
acknowledge fire issues. By contrast, the trend in ratings for photo SA is very different,
reflecting the preference amongst urban inhabitants for forest environments suitable for
recreational activities.

1
Deep rural Developing rural Outer urban Inner urban

SC SA SD SB SE

Figure 5.8. Public preference for stand structure by municipality group in terms of attractiveness

1
Deep rural Developing rural Outer urban Inner urban

SC SA SD SB SE

Figure 5.9. Public preference for stand structure by municipality group in terms of management
criteria
150
Overall, the results presented in this section show that there are differences in ratings of
forest characteristics across both user groups and the urban-rural gradient. This implies a
positive answer to the second question posed in Box 5.1, namely that there are differences
in preferences. It is also worth noting that the survey was conducted in an area which lacks
a major urban centre and consequently in a region with such a feature the contrasts could
be even greater. The next section explores the third research question which asks whether
there are indications that human preferences might threaten the ecology of forest
ecosystems.

5.3.3. Public preferences for forests and ecology of forests ecosystems

The results presented earlier in Table 5.1 indicated that photo SA (an even stand forest
with no understorey) got the best rating for management, while the type of forest shown in
photo SC (even stand with understory) was regarded as “not good management”. This
implies that the public associates removal of ground vegetation with good management
practices. Such an interpretation is supported by the fact that the burnt forest photos (FE
and SE) got better ratings for management than attractiveness criteria. Controlled fire
techniques have been used in countries with high wild fire risk (Portugal, Spain, Greece) as
a tool to remove understorey vegetation and slow fire spread. However, the positive effects
of scrubland ground cover in reducing soil erosion and promoting water retention are well
known by forestry practitioners. There are also well known beneficial effects of continuous
cover forestry, but in this survey photos of a contiguous forest (FC) got better ratings for
attractiveness rather than management.

The results of this study therefore provide a further example illustrating the tensions
between aesthetic and ecological objectives in multifunctional landscape planning.
Management practices that can be used to reduce a threat such as wild fires might also
affect the ability of forest to provide other ecosystem services such as soil and water
protection. This has implications for the issues discussed in Chapter 4. On a catchment
scale it could also mean that trying to incorporate public preferences for forest
management could complicate effective partnerships between municipalities located in the
top and downstream parts of a watershed.

5.3.4. Contrasting verbal and visual approaches

Results from the section of the survey where respondents were asked verbally to select
forest characteristics are summarised in Table 5.6. Overall, 52% of the interviewees
considered a continuous forest to be more attractive than a patchy forest, while 45.3%
preferred the latter. 2.7% did not answer the question. When management was considered
151
42.4% of respondents considered connectivity of forests as a good management practice
while 57.6% associated good management practices with discontinuity of tree cover thus
choosing fragmented forests. These results indicate that there was a relatively small change
in opinion when respondents were asked to consider attractiveness or management criteria.

In responding to a verbal question regarding stand structure preferences 81% of

interviewees said that they preferred uneven stands in terms of attractiveness. When asked
to consider this choice on management grounds the proportion barely changed (see Table
5.6).

Table 5.6. Verbal stated preferences for contiguity of forest cover and stand structure

Attractiveness Management

Connected 52% 42.4%

Fragmented 45.3% 57.6%

Even aged 18.7% 18.7%

stands

Uneven 80.8% 81.3%

stands

Note: For the attractiveness question there were 10 and 2 missing values for fragmentation
and stand structure respectively.

Figure 5.10 plots the mean ranks of four photos (excluding the burnt area FE) showing
different levels of fragmentation differentiating between respondents according to their
stated verbal preferences. Very little difference is apparent. A similar situation occurred
when verbal and visual approaches were compared regarding preferences for stand
structures (Figure 5.11).

Visual preferences attractiveness

most attractive 5 the

5
mean value 1 the

least attractive

4
3
2
1
FC FA FD FB
gradient of fragmentation connected to fragmented

CONNECTED FRAGMENTED

Figure 5.10. Mean rank of attractiveness for photos (visual approach) according to verbal stated
preferences for percentage of forest cover

152
 Visual preferences attractiveness

most attractive 5 the

5

mean value 1 the

least attractive
4

1
SC SA SD SB
gradient of simplification even to non even stands

EVEN AGED UNEVEN STANDS

Figure 5.11. Mean rank of attractiveness of photos (visual approach) according to verbal stated
preferences for stand structure

The results in Table 5.6 indicate that there were only small changes in public preferences
when they were asked about attractiveness and management criteria verbally. Referring to
the results obtained through the use of photos, the Spearman rank correlations show that
the majority of the public ranked the set of photos in the same way for the two criteria
under analysis. Therefore the verbal and visual approaches delivered similar overall results
in the sense that both analysis suggest that the public do not make a distinction between
attractiveness and management criteria.

Nevertheless, the results in Figure 5.10 show that independent of respondent‟s stated
verbal preferences for either a “patchy” or a “contiguous” forest the photos showing
different levels of forest cover were similarly ordered. The same happened in the case of
stand structure, as shown in Figure 5.11. This result implies some inconsistency between
responses to verbal and visual questions.

5.4. Discussion

This study has focused on the comparison of public opinions regarding percentage of forest
cover and stand structure given attractiveness and management objectives. A Spearman
rank correlation between the rankings of the two sets of photos on attractiveness and
management grounds showed that the great majority of respondents (around 80% when
photos of burnt forests were removed from the study) ordered the photos in the same way
irrespective of the criterion under consideration. Including the photos of burnt scenes (FE
and SE) in the calculations further increased the extent of positive correlations,
highlighting the value of sensitivity analysis when working with visual images (Palmer and
Hoffman, 2001).
153
Independent of whether the photos of burnt scenes were included in the analysis, in the
vast majority of cases where higher (lower) rankings were given for attractiveness then
similar ratings were given for management. This implies that whatever was considered by
the public as “good management” was also regarded as aesthetically pleasing, while poor
management was not aesthetically pleasing. These results therefore support the view that
the general public is not able to disaggregate “the eye from the beholder”. This suggests an
avenue for further research where transdisciplinary approaches “work” to create public
awareness of the importance of multifunctional forest management. In my opinion,
research needs to attempt to distinguish between stakeholder‟s attractiveness and
management preferences for forests instead of simply associating public preferences with
interviewees‟ background.

Differences in preferences for forests across user groups are widely reported in the
literature (Ribe, 1989; Roovers et al., 2002). In this study different aesthetic and
management preferences for forests were found between types of user (e.g. multi and
single users) as well as between rural or urban populations. This accords with previous
findings that urban and rural dwellers have different needs; urban residents‟ favouring
contact with nature, attractive environments, places for recreation, play and privacy
(Matsuoka and Kaplan, 2008). On the other hand, rural residents value collection of
products (e.g. fuelwood) and a natural environment that is able to promote socio-economic
wellbeing, one example of which is tourism (Rogge et al., 2007a).

The association between aesthetic preferences and respondent background and place of
residence does not, however, negate the advantages in surveying both aesthetic and
management opinions. Firstly, although it is likely that aesthetic and management
preferences will be positively correlated (as was the case for 80% of respondents in this
study) it may still be the case that some respondents make a distinction (18% of
respondents had an inverse relationship between aesthetic and management preferences).
Second, by contrasting attractiveness and management preferences for forests the research
has highlighted how simply following aesthetic preferences may have negative
consequences for ecosystem functions. It is therefore possible that a better understanding
of human-nature relationships can be achieved if attractiveness and management criteria
are addressed separately.

A key issue, consequently, is how to integrate the different preferences of diverse users
into forest management in a way that human preferences do not threaten the ecology of
forest ecosystems. This calls for active involvement of the public in forest management . It
is clear that the public needs to be engaged in order to understand the trade-offs that need
154
to be made between human preferences (both aesthetic and management) and the ecology
of forest ecosystems (Fry, 2001; Gobster et al., 2007). Creating more sustainable
landscapes, in which forests may have an important role by providing services such as soil
and water protection, will depend mostly on the farmers and forest landowners who in
countries such as Portugal and Spain are the owners of the majority of the forest land.
Furthermore, in countries with problems of wild fires, the public in general needs to be
aware that good management does not necessarily imply the complete removal of
understorey from the forests. There is here scope for further engagement with the public to
attain sustainable forestry management. General public and forest stakeholders need to
fully embrace the challenge of multifunctionality (Brandt et al., 2000; Fry, 2001; Mander
et al., 2007; Selman, 2002) in which there is a need to conciliate forest management for a
variety of goods and services. Completely removing understorey (or at least planning to do
so) from the forests will undermine the multifunctional management strategy that was
proposed in Chapter 4.

The fourth research question in this study focussed on contrasting verbal and visual
approaches. It was found that verbally stated preferences did not consistently correspond
with the rankings of photos (visual preference). This is a similar finding to the work by
Tahvanainen et al (2001). One of the reasons for such a result is that respondents may have
had different mental images of “continuous forest” and “patchy forest” in the case of
fragmentation and “even stands” and “uneven stands” in the case of stand structure. This
may explain why the ratings of the photos were similar independent of stated verbal
preferences. In essence, respondents had different mental images of those concepts that
were “made concrete” by the use of the photos. For example, when only making use of
verbal data it was not possible to depict a trend in public preferences for percentage of
forest cover. However, by using the photos public preference for a certain degree of
discontinuity on forest cover was apparent (photo FA). This puts emphasis on the need to
widely use visual tools in questionnaire surveys (Tahvanainen et al., 2001). Image capture
technology along with simple photo manipulation was found to be very effective as a
means of investigating preferences in this study.

Despite the efforts made in eliminating from the photos other factors likely to distract
respondents from the factors under analysis (e.g. only varying forest cover and stand
structures) it was apparent during the survey that many people were trying to locate the
photos within the study area. It is therefore possible that an association was made between
the characteristics shown and the places in which the forests were located. For example,
continuous forest cover such as shown in photo FC is typical of the region of the Geres
155
National Park as so when respondents were rating the photo may have made this
connection and this, in turn, influenced their ratings for the two criteria being considered.
Therefore, although there were advantages in showing forests from the study area it needs
to be acknowledged that this could have influenced the results.

5.5. Conclusions

In addition to distinguishing between preferences for forests across user groups this study
has shown that not only aesthetic preferences but also preferences for management strategy
can compromise the functioning of forest ecosystems. Consequently, public opinion
regarding management also needs to be included in the aesthetics-ecology debate. There is
a need to create public awareness of the impact their preferences might have on the
ecology and thus contribute positively to the nature-society and aesthetic-ecology debates.
By contrasting attractiveness and management criteria within forest management this
research has presented a simple approach to engage the public in the “aesthetics-ecology”
(Gobster et al., 2007) or “nature- society” debates (Selman, 2006). Methods such as the
one employed in this study in which public is engaged in exercises of comparing several
criteria for forestry management could well help in such a demanding task.

156
 “If you have seen one rural
place, you have seen one rural
place” (OECD, 2006: 36)

CHAPTER 6. GOVERNING FOR SUSTAINABILITY:

IMPLEMENTING SUSTAINABLE FORESTRY MANAGEMENT IN
NORTHERN PORTUGAL

157
Abstract

Managing forests in a way that user groups, sustainability practitioners and forestry
institutions all agree with is not easy. For agreement to occur, via viable landscape design
procedures, through which multiple functions (production, environmental protection and
recreation) may be coordinated by means of innovative planning, there is a need to
negotiate a set of common objectives and shared responsibilities. This chapter explores the
policy dimensions of multifunctional forest management, and, through an exploratory case
study, proposes a mechanism for cooperative planning and institutional design. The case
study involved two communities in the Minho region of Portugal (Gavieira and Entre
Ambos-os-Rios) combining the local communities, the National Park, and local forestry
and tourism offices. The case study created, developed and validated two scenario
storylines through a series of participatory processes (two focus groups meetings, one
comprehensive workshop and one expert meeting). One scenario focussed on continuity of
the traditional management patterns, with an emphasis on direct goods such as timber and
livestock grazing (traditional multifunctionality). The other concentrated on indirect
ecological services, such as soil and water protection, as well as carbon sequestration
(new-multifunctionality). An attempt was also made to implement the storylines through
initiating a pilot project in both of the case study areas. However, there were neither robust
planning mechanisms nor adaptive governance systems with the capacity to put into place
forest management “futures” likely to deliver more sustainable landscape-scale uses in
these areas. This chapter illustrates the difficulties in forging governance systems that have
the capacity and the vision to be able to put sustainable development concepts into
practice, given a policy setting that is confused, contradictory, and where the “status quo”
tends to be given prominence.

Keywords: multifunctional forests, integrated landscape planning, sustainable forest

management, rural diversity in Portugal

158
6.1. Introduction

Planning and management for sustainable use of natural resources enforced by governance
systems able to deal the challenges involved are seen as key vehicles for a transition to
sustainability (OECD, 2002, 2006). According to the European Landscape Convention,
“planning” is a forward-looking action to enhance, restore or create landscapes (Council of
Europe, 2000), while “management” means the routine tasks needed in order to achieve the
planning goals. Planning is generally associated with integration across space; it is a “place
making” instrument. Recently, there has been a trend to broaden the concept emphasising
the integration of place within the whole landscape by means of a spatial planning
approach. However, the mere act of establishing goals and adopting programs will never
be successful unless such objectives are made fully operational and enforceable (Carlman,
2005). In addition to effective planning, there is a need to put into practice effective
governance systems for sustainability (Adger et al., 2004; CouncilofEurope, 2000; OECD,
2002).

Governance over a territory can be understood as “the emergence and implementation of

innovative shared forms of planning and managing of socio-spatial dynamics” (CEMAT,
2007:29). To govern a territory means to negotiate a set of commonly agreed objectives as
well as a framework of shared responsibilities by the use of spatial development strategies
and policies (CEMAT, 2007), thus calling for a whole landscape approach. A “whole
landscape approach” is defined by Dolman et al, (2001:306) as “a process of integrated
planning across property boundaries that optimizes the amount, location, configuration
and management of...landscape elements”. Despite increasing recognition of a need to
manage the whole landscape (Dolman et al., 2001) “traditional” planning is still embedded
in sectoral approaches (Selman, 2002). Consequently, there are landscape ecological plans
for agriculture, forestry and coastal areas where the link between these and the whole
landscape (including living communities) is vague. In addition, there are landscape
economic plans, landscape social plans, landscape heritage plans, and so on. (Selman,
2002) A whole landscape approach requires a full integration of responsibilities and
“mind-sets” between planning institutions (Dolman et al., 2001).

Such institutions in Western Europe have, or at least are supposed to have, higher abilities
to deal with challenges to attain sustainability (O' Riordan and Stoll-Kleemann, 2002;
O'Riordan and Voisey, 1998). Western European countries have well established public
administration systems based on rule of law, rational specialisation of tasks, together with
a professional and high skilled civil service (OECD, 2002). Yet, the extent to which these

159
“well-functioning institutions” and the operational plans they put into practice, really do
promote sustainable development is now being critically examined. For example, there has
been a recent call for new forms of cooperation between public and private property rights,
involving “fresh” institutional arrangements and funding for an orderly transition to more
sustainable landscapes to occur (Milligan et al., 2009; Nicholson-Cole and O'Riordan,
2009). It has also been suggested that a “new paradigm” for policy and governance is
crucial for any successful move forward (OECD, 2006).

This chapter explores the issues related to innovative planning and governance for
sustainable forestry management in rural areas of Northern Portugal. It explores the ways
in which forests can be integrated into the whole landscape, for community living as a way
to move towards sustainability. In Portugal, as in other European countries, forestry
institutions have offered financial support both for new plantations and for protection of
established forest areas (DGRF, 2007b). A review of the different ways in which European
countries deal with the forestry sector is presented in the work by Marey-Perez and
Rodriguez-Vicente (2008). Despite considerable criticism of the current performance by
top-down approaches, in economic and regional development strategies currently being
pursued EU countries have encouraged reforestation (Okkonen, 2008). The area of forest
in Europe has increased by almost 13 million ha in the past 15 years mainly due to planting
of new forests and natural expansion of forests onto former agricultural land (MCPFE,
2007; Nabuurs et al., 2001; Niskanen and Lin, 2001). This reversion of deforestation rates,
and even an increase in area of forests, is explained by the forest transition concept
(Mather and Needle, 1998). There are studies reporting forest transition in countries such
as France (Mather et al., 1999), Spain (Marey-Pérez and Rodríiguez-Vicente, 2008),
Scotland (Mather, 2004) and Denmark (Mather et al., 1998). Mather (1992) summarises
the overall trend as a change from a pre-industrial stage through an industrial stage towards
a post-industrial stage of forestry. While the first two stages can be ascribed to traditional
utilitarian management approaches, forest management in the post-industrial stage
perceives forests less as commodities and more as functioning ecosystems serving multiple
purposes. The post-industrial stage calls for multi-purpose forestry in which forests provide
a wide range of functions across whole landscapes (Mander et al., 2007; Vejre et al., 2007).

The concept of multifunctionality, either addressed throughout a single land use type such
as forestry or agriculture, or by means of integrating of several land use types
(multifunctional land use), is of major importance in context of rural development (Dwyer,
2007; Hagedorn, 2007). The evidence in rural areas of Europe, both from the development
of various scenarios, as well as from case-studies promoted by programs such as FAIR
160
(from which one example is the Impact project (Knickel and Renting, 2000), reveal the
contours of a new development trajectory in which the key features are diversity and
multifunctionality. Diversity is reflected in the actors involved, the particular activities
undertaken, and the patterns of motivation that emerge (Elands et al., 2004; Elands and
Praestholm, 2008). Multifunctionality is the simultaneous and interrelated provision of
different functions from a single land use type (Mander et al., 2007). The concept of
multifunctionality has attracted the attention of several scholars (Brandt et al., 2000; Fry,
2001; Mander et al., 2007; Selman, 2002, 2006; Vejre et al., 2007). An array of topics
exploring the concept have been presented in various international conferences such as
“multifunctional landscapes” (Brandt et al., 2000). Multifunctionality lies within the
operational role of sustainable development and is based on the assumption that
agricultural and forestry usage have always fulfilled more than just their primary aim of
producing food, fibre, timber and fuel (Knickel and Renting, 2000; Mander et al., 2007).
Hence, the study of “viable multifunctionalities” aiming at creating sustainable livelihoods
in otherwise deprived rural areas is of upmost importance (Pinto-Correia and Breman,
2008; Selman, 2002).

Multifunctional land use can be achieved in three ways: 1) by pursuing different goals in a
corresponding mixture of separate land use types, 2) by pursuing different goals on the
same parcel of land, but sequentially in time, or 3) by integrating from the beginning and
coordinating the different goals to accomplish them simultaneously (de Blust and Olmen,
2005). The first way can be defined as spatial multifunctionality because different spatial
units (land use types) have clearly defined management goals. In spatial multifunctionality
each piece of land has one function, but when zooming in or out to see a full landscape, it
appears to be more or less multifunctional (de Blust and Olmen, 2005). By contrast, in the
second and third variants, different goals are attained in the same spatial unit (successively
in the second and simultaneously in the third). This type of multifunctional land use can be
defined as integrated multifunctionality (Brandt et al., 2000).

Forestry is increasingly recognised as a critical activity for retaining viable livelihoods

within rural development for many EU countries (Slee, 2006, 2007b; Slee et al., 2004;
Stengera et al., 2009). Rural development is recurrently addressed within two contrasting
perspectives namely, endogenous and exogenous development. The former is conceived as
a bottom-up process in which rural development results from local initiatives. The latter is
conceived as a top-down process in which rural development is the result of political and
economic dynamics created outside rural areas (Okkonen, 2008). These different processes
of development raise different rural discourses. Research by Elands and Wiersum (2001)
161
distinguishes between different roles for forestry based on five discourses of rural
development. These are agri-ruralist, hedonist, utilitarian, community stability and nature
conservation.

Within the agri-ruralist discourse, forests are regarded as complementary to agriculture,

and hence, should not become too dominant in any given area thereby jeopardising
farmers‟ traditional stewardship. The hedonist discourse promotes the creation of
“wilderness” areas and places forestry as a driver for recreational attractiveness. Under the
utilitarian discourse forests are mostly seen as a source of income. In the community
stability discourse, forests are planned to sustain forest dependent communities. This
variant regards forests as an income provider, either by optimising forest labour
employment or by optimising forest production as a complement to farm production. In the
nature conservation discourse, intrinsic protection of forests is an aim in itself, hence near-
to-nature management practices are stimulated. In theory, boundaries between these
various forest management discourses are discernible. However, the extent to which it is
possible to distinguish between these various interpretations in practical terms is more
problematic. It therefore seems sensible to explore the relationships between these
discourses and the type of multifunctional land uses that they are likely to create.

Because there are complex ecological, socio-cultural and economic drivers affecting the
future(s) of rural Europe, the development and assessment of land use scenarios can play
an important role in promoting the understanding of such complex and uncertain systems
(Shearer, 2005; Tress and Tress, 2003). Kok et al (2006:264) defined scenarios as
“plausible, challenging and relevant stories about how the future might unfold that can be
told in both words and numbers”. There are several scenario studies addressing the
possible futures of rural Europe. Some examples are projects such as EU Ruralis, Visions,
Prelude, and Corason (Tovey, 2008). Some are multi-scale (global and local scenarios)
such as the work developed in the Med Action Programme by Kok et al. (2006) for
Europe, Mediterranean and watersheds within the Mediterranean. Land use suitability
analysis is a widely used approach for modelling future land use allocations in scenario
studies. The aim of land use suitability analysis is to identify the best site for some activity
given a set of potential sites (Malczewski, 2004). In this type of analysis relevant attributes
of the “object” to model are known (e.g. ecological thresholds for tree species). By
performing both land suitability analysis and land use allocation, the alternative sites are
ranked based on their characteristics so that the best site can be identified for a specific
land use (Malczewski, 2004). Another common feature of the studies dealing with
scenario development across Europe is the use of participatory processes (Kok et al.,
162
2006). Despite wide agreement on the advantages of participatory processes, there are also
shortcomings related to the effects of “powerful” stakeholders that may clearly influence
the participatory process. A review of participatory processes and their possible drawbacks
is offered by Milligan and O‟Riordan (2007). These relate to issues such as the credibility
of the policy framework for adapting to the demands and expectations of stakeholders,
representativeness of participants, time for resolving conflicts, shifts in policy and
economic conditions as the process evolves, and means of ensuring wider interests are
fully taken into account. These are not simple conditions to meet (Milligan and O'Riordan,
2007).

In countries such as Portugal during the recent past (1950s) forests in rural areas were
regarded as an agro-silvo-system where agriculture, animal husbandry and forests
complemented each other (Pereira et al., 2005; Pereira and Fonseca, 2003). One example is
the use of understorey from forests as a bed for livestock which in turn provided manure to
fertilize the fields. This was reported by several authors (Firmino, 1999) as a form of
“sustainable” multifunctionality connecting agriculture, forest and grazing area land use
types. This traditional management system worked well in a localised economy context.
Successive Common Agriculture Policies (CAP) are blamed for disrupting these traditional
systems by incentivising mass agricultural production against which such “traditional
management” practices could not compete (Andresen and Castelbranco, 1993). As a
consequence of the disruption of the equilibrium between the mosaics of agriculture,
forestry and grazing areas, the traditional management which created an integrated
multifunctionality is no longer working, raising issues such as depopulation and ageing of
rural population in many inland rural areas. The study of “viable” multifunctionalities in a
post-modern Portugal is of major environmental, social and economic importance (Pinto-
Correia and Breman, 2008).

By using scenario approaches in two specific case studies this study focused on: 1)
developing scenario storylines for integrating forestry within the whole landscape mosaic
in rural areas of Northern Portugal, 2) exploring “viable” designs for land use as well as
planning mechanisms to implement those designs, 3) examining appropriate governance
systems likely to implement the storylines seen as “viable” and, 4) implementing those
storylines through a pilot project. The research particularly addresses the issues that
planning and governance systems have to tackle in order to place multifunctional forests
within the whole landscape mosaic as a means to move towards effective sustainability in
revitalising rural areas of Northern Portugal. The four major questions addressed here are
presented in Box 6.1.
163
Box 6.1. Questions addressed in the case studies

1. What scenario storylines for sustainable forests and rural communities are
considered viable by local stakeholders in northern Portugal?
2. What type of multifunctionality (ies) are likely to be viable?
3. What type(s) planning approaches as well as governance systems are needed
to implement those multifunctionalities?
4. How can these two storylines be implemented through a pilot- scheme?

6. 2. Study area and methods

6.2.1 Study area and case study selection

The Minho region is located 38 km north of Porto and stretches to the frontier of Galicia in
north-western Spain. Within Minho there are two sub-regions: the upper area (Alto Minho)
and the lower area closer to Porto (Baixo Minho). Arcos de Valdevez and Ponte da Barca
are two Municipalities located in the Alto Minho region of Portugal. In this study
storylines for “viable futures” were created for two parishes, namely Gavieira and Entre
Ambos-os-Rios in Arcos de Valdevez and Ponte da Barca municipalities, respectively.
Important changes in agro ecosystem resources in the Minho region were reported in the
Millennium Ecosystems Assessment programme (MEA, 2003; Pereira et al., 2005). In the
Minho region local communities recognized that local forests provide direct benefits which
increase social wellbeing such as wood, fuel and fodder. In addition, forests contribute to
local income generation throughout the trade of timber. Local communities also recognized
indirect services provided by forests such as air purification, healthy environment, food for
hunted animals, water springs and oxygen (Pereira et al., 2005).

The selection of the parishes in which this research took place was one of the outcomes of
a questionnaire survey undertaken in previous research. In 2007 fourteen municipalities
within the Lima and Cávado watersheds in Minho region were surveyed in order to study
public uses and preferences for forests (Chapters 4 and 5). During this survey fourteen
municipal forestry offices in the Minho region were contacted by the author in order to
gather information regarding the dynamics of the forestry sector in each municipality.
From this initial round of meetings two municipalities within Lima watershed, namely
Arcos de Valdevez and Ponte da Barca, were selected due to their proactive attitude
towards sustainable land use planning, and their enthusiasm to put into practice a
participatory process for forest scenario development. These two municipalities voluntarily
embraced the challenge of scenarios development. Their forestry offices assumed the
responsibility of contacting relevant stakeholders and arranging local meetings. The
164
organizing leaders of the group were the rural development association ARDAL in Arcos
de Valdevez, and the municipal forestry office in Ponte da Barca. Those two entities
drafted an invitation list of relevant stakeholders to be involved in the focus group
meetings in each municipality. The participants‟ selection was done jointly by the research
team and local leaders. This was an agreed arrangement as it was considered vital to obtain
both representativeness and credible participation to work through the offices of the
regional and local governmental machinery. Respondents were therefore selected for their
knowledge, viewpoints, occupation and degree of involvement in forest management. This
approach is also endorsed by Milligan et al (2009) in their work on coastal futures planning
in North Norfolk. In Arcos de Valdevez, the relevant parties suggested inviting national
park officials, local farmers, tourism and industry stakeholders, in addition to the ARDAL
and municipal forestry office. In Ponte da Barca the stakeholder group included farmers,
local fire-fighters, and local government decision makers, as well as the forestry engineer
from the municipal forestry office.

6.2.2. Forestry and land management in the study area

Forests are a prominent land cover type in Minho region (Moreira et al., 2001a; Moreira et
al., 2001b; Pereira et al., 2005). Arcos de Valdevez and Ponte da Barca are included in the
Alto Minho Regional Forestry Plan “Plano Regional de Ordenamento Florestal PROF Alto
Minho”. This regional plan refers to the Forestry Policy Act (1996), which provides the
national strategy for forests in Portugal, as well as to the Plan for the Sustainable
Development of the Portuguese Forest PSDPF (1999). Also at national level there is a plan
to protect forests against fire (PNDFCI). The relevant instrument in the Portuguese
constitution states that “the state will promote forestry policies according to ecologic and
social circumstances” (Portuguese constitution, 93rd article, number two). At national level
there is also a funding scheme created from revenues from petrol consumption (Fundo
Florestal Permanente) which provides financial support for forestry related investments
(DGRF, 2007b).

Regionally, in addition to PROFs (forestry sector plans), there are the PROTs (Regional
Plan that regulates all land uses), and the PEOT (created exclusively for regulation of land
use allocation in protected areas). All the three regional plans are only mandatory on the
public land which represents approximately 2 % of all forest land. Plans that are mandatory
at local level for private and communal property are the PMOTs, which include the
municipal director plan (PDM). This regulates all land uses, the urbanization plan (PU),
and other specific plans (PP). In the case of private property included in protected areas,

165
there is a special plan, PEOT, that operates throughout the management tiers. For the
protected area, the PO regulates all land uses. At local level, there are landscape plans
called Plano Director Municipal (PDM) which incorporate the municipal plan for defence
of forests against fire (PMDFCI). Table 2 shows the plans and the scale addressed. These
are the official plans that, in 2008, controlled the forestry sector. There have been,
however, continuous changes in the “strategy” for forestry polices in Portugal. From the
1980s until 2004 the state financed private forestry organizations in order to entice both
private and communal landowners. From 2004 onwards there was a shift from financing
private forestry associations to the direct support of municipal forestry offices.

Table 6.2. Current arrangements for planning and funding forest management in Portugal

Scope Legal document Law/plan Regions

International Portuguese constitution N155 12 08 2005 Portuguese territory

and
National Plan for “ordenamento do PNOT
National
territorio”
Homogeneous
Forestry Policy Act (1996) region
Law forestry policy
Plan for the Sustainable Development
of the Portuguese Forest PSDPF
(1999) EFN
National plan for defence of forests PNDFCI
against fire
Plan for combat desertification PNCD

Regional Regional plan for ordenamento do PROT Sub region

territorio
Special plan for ordenamento do
PEOT
territorio (protected areas)
Regional plan Ordenamento florestal
(forests) PROF

Municipal Municiplal plan ordenamneto do PMOT Land use classes

territorio
Municipal directive plan
PDM Land use classes

Municipal plan for defence of forests

PMDFCI Sector
against fire

Local Management plan for private forests PGF Sections/ forestry

project

Management plan for communal PUBs

forests
Management plan for private and
ZIFs
communal forests

166
6.2.3. Creation and development of scenario storylines

The first contact with the municipalities in Minho region occurred in early 2006 with the
aim of gathering socio-economic and digital data. In 2007 a questionnaire survey was
conducted in fourteen municipalities in order to survey public uses and preferences for
different qualities of forest landscape. The creation and development of the storylines
occurred in two separate sets of focus group sessions between January and May 2008 in
the two parishes. During a combined workshop in May, the two groups together with local
forestry institutions validated the storylines. The workshop aimed to further discuss the
storylines as well as sharing the experiences of the two neighbouring parishes. Because
high level representatives were present in the final workshop the focus group members had
the opportunity of discussing their views with the Peneda-Geres National Park manager.
An expert meeting was held in July 2008 in order to study the possibility of
implementation of the two scenarios through a pilot-scheme.

During the two focus group meetings storylines were created and developed, aided by land
use modelling tools such as multi-criteria evaluation, in order to create landscape
suitability maps and different allocations land use types for the two contrasting storylines.
In the workshop and expert meeting those storylines were validated. During this stage
photo-montages were created to help stakeholders visualise the outcomes of the two
scenarios. The scheme in Figure 6.2 summarises research goals, objectives and phases.

167
 Goal:

Policy dimensions for the Objectives

implementation of
multifunctional forest 1. Create storylines 2. Test their
management in rural areas for “viable futures” implementation

Questionnaire survey Focus groups and final Expert meeting

workshop
1) Setting the scene for Economic
scenarios development 1.Creation and evaluation
development of storylines
2) Choice of the study sites Piloting
2. Validation of storylines implementation
and explore ways for
of scenarios?
possible implementation

Policy dimensions
Patterns of planning guidelines, financing arrangements, local to regional
political relationships, landscape histories, landscape ownership and
neighbouring cooperation

Framework for implementing multifunctional forest management in rural

areas of Northern Portugal based on broad sustainability principles

Figure 6.2. The Structure of the research approach

168
6.2.3.1. Focus group meetings

The scenarios development phase occurred between January and March 2008. Two focus
group meetings were held in each one of the two parishes (Gavieira and Entre Ambos-os-
Rios). The overarching goal for the two focus group meetings was formulated on the basis
of three questions. These were: “How can forests in this region: 1) fulfil society‟s needs; 2)
be economically viable; and 3) be ecologically rich (e.g. biodiversity was explained as a
high diversity of animal and plant species).

In January 2008, during the first focus group meeting, three exercises were conducted. In
an icebreaking exercise, participants ranked two sets of photos of forested landscapes
according to scenic beauty and management criteria. This process was based on established
research practice in forest aesthetics (Ribe, 1989; Sheppard and Harshaw, 2000). In the
second exercise, individually, and in groups, the participants addressed their feelings about
the forests of the region focusing on the following feelings: a) PROUD (what makes feel
you proud of the forests of this region?); b) SADNESS (what aspect of the forests causes
you sadness?); c) HOPE (what hopes do you have for the forests of this region?) and
finally d) FEARS (what do you fear that can happen to the forests here?). This is a variant
of the well known strengths, weaknesses, opportunities and threats (SWOT) used in many
assessment exercises (Leskinen et al., 2006). The usage of these simpler words had more
resonance for participants who were not schooled in management speak. A secondary task
was to enumerate three characteristics that a “good” forest in the region should have.
Participants had 5 minutes to write down their own ideas and in a subsequent 15 minute
discussion the individual ideas were grouped into clusters to guide further discussion.

The last exercise introduced the theme of scenarios defined as meaning “possible futures”
for the region. This exercise was explained as no more than “explore plausible futures” that
could be seen as likely to occur, without being intended to be predictions. On the basis of
previous work (Chapter 4) two scenario storylines were proposed as a basis for analysis.
These were “carbon offsetting forestry” and “conventional forestry” scenarios. In the
former, carbon sequestration and energy from forests biomass were supposed to take over
all the other “old” uses. In the latter these “new” uses were implemented on a small scale
based on the view that reluctant farmers are likely to use the precautionary principle in
adopting new uses. The two baseline storylines were proposed based on the results of the
questionnaire survey conducted in 2007. According to these results forestry biomass and
carbon sequestration were regarded as new uses likely to have a major impact on the future
of the region (Carvalho-Ribeiro and Lovett, 2009). These baseline storylines were
discussed for twenty minutes. In the last 10 minutes the group evaluated the session.
169
The second focus group meeting (March 2008) started with a brief presentation of the
results of the January meeting. During the following group discussion, the participants
gave their opinion about the results presented. New versions of the storylines were created
in order to incorporate participants‟ proposals. This presentation included the first version
of the land use models that were created for each scenario (Section 6.3.1). The discussion
around the questions raised (see Box 6.3 below) further contributed to the storyline
development. As can be seen in Box 6.3 participants were given the opportunity to
incorporate further ideas related to the “construction” of a third distinct storyline. In a final
stage in the discussion, issues of implementation for each scenario were addressed; these
included the institutional arrangements regarded as likely to promote each one of the
scenarios created.

The aim of the focus groups was to explore how two differently selected, but
representative, groups of interested and knowledgeable parties addressed the concept and
image of a sustainable managed forest in their region. The two focus groups were
conducted in two study areas so as to capture the possible differences in perception and
expectation.

170
Box 6.3. The structure of questions used in the focus groups in each of the two study sites.

Focus Group 1 Focus Group 2

(January 2008) (March 2008)

Aims

“1) Build cohesive working groups, 2) Develop the storylines drafted on the
study of people‟ opinion about a “good” previous focus group meeting
forest 3) introduce the topic of scenarios as
“possible futures”

Questions

Q1: What is an attractive/well managed
forest in this region?
Q2: What are you feelings in relationship of
the forests of your region?
Q3: What type of forests do you would like
to have in this region?
Q4: What are the possible futures for the
forests of this Municipality (20 minutes)
Q1: In your personal opinion how can the
storylines be improved
Q2: Are these storylines credible?
Q3: Are there other(s) possible futures?
Q4: What is the most desirable future for
this parish?
Q5: Who will win/lose in these futures?
Q6: What stop us to have the desirable
future?
Q7: What type(s) institutional arrangements
will promote the desirable future?

Exercises

Q1:Ranking a set of photos (5 minutes) All questions:

Q2: Fill the PROUD/SADNESS and
HOPE/FEAR boxes (20minutes)
Q3: “Good” forest characteristics (20
minutes)
Q4: scenarios (25 minutes)
Evaluation
Each participant wrote down their
individual
Ideas (approximately 3 minutes) this was
followed by a group discussion. At the end
of the session a summary of the results for
each question was made.
Evaluation

6.2.3.2. Land use change models

Land use suitability analysis aims at identifying the most appropriate spatial pattern for
future land uses according to given suitability criteria (Malczewski, 2004). Based on
climatic data such as temperature and rainfall as well as ecological variables (Table 1,
Annexe 5), suitability maps were created for broadleaves, coniferous, eucalyptus,

171
agriculture and pastures by using multi-criteria evaluation techniques. All these base maps
were downloaded from the Portuguese Atlas available at (www.atalsambiente.pt). From
the national data base, regional maps for the study area were extracted with the following
coordinates: minimum x=168 000, maximum x 205 000, minimum y= 530 000, maximum
y= 570 000. The base maps had a 20 meter pixel size and the number of columns and rows
was 1850 and 2000 respectively.

Table 6.4. Constraints and factors in the suitability maps

Ecological Coniferous Eucalyptus Broadleaves Agriculture Pastures

requirements (maritime pine) globulus (common oak) (e.g.corn) (natural)

Base maps used Urban=no, unproductive=no and water bodies=no.

Constraints

Factors: Max 45 Max 25 Max 45 Max 15 Max 30

Slope (degrees)

Frost (n. Cope well with Very sensible Sensible but all Very sensible Cope well with
months) less frost for more (less than one region meets its (less than 1 frost for more
than 1, to than 4 months month) requirements month) than 4 months
between 3 and 4

Ecological Adapted to all All, All Mediterranean All

zoning
Mediterranean
Atantic

Insolation (n. Tolerate all Max number Tolerate all Max number Tolerate all
hours of hours possible hours possible
sunlight) 1800,
to 2500

Mean Annual All Copes with all Grows better All faster All
Temperature faster growths between 12.5 growths
(degrees between 12.5 and 15 but still between 12.5
Celsius) min 7.5 and 15 does well across and 15
to max 15 the region

Soil type Faster growth in More sensible to Favours More sensible All
Rankers, cambisol but type of soil cambisol but to type of soil
Cambisol still grows well prefer cambisol still grown in prefer cambisol
humicos in rankers rankers

Distance to Can be far way Consumes a lot As close as As close as Cope well in
rivers (meters) of water. Place it possible possible dry places
far ways from
the rivers

Distance to Not Not surrounding As close as As close as Away from

urban (meters) surrounding urban areas but possible possible urban
urban areas but relatively close
relatively close (10 km)
(5 km)

Distance to the As close as Close As close as As close as As close as

same Land use possible possible possible possible
(meters)

Source: Characterisation of the requirements of tree species in Portugal see Annexe 5,

Table 1 for detailed description
172
The land use modelling criteria, as well as the weightings for factors (Table 2, Annexe 5),
were discussed with the parishes representatives on the January and March focus groups
outlined in Box 6.3. The land use models were revised after the workshop and the expert
meeting. Subsequently, these suitability maps were combined through a multi-objective
land use allocation command in the IDRISI ANDES GIS (Eastman, 2006) to produce land-
use allocations under the different storylines (Table 3, Annexe 5). The land use allocation
was made based on the description of scenario storylines (Table 6.7- row showing land use
in 100 ha).

Based on the description of the scenario storylines (Table 6.7) land use allocations were
made according to the values in Table 6.5. Detailed information about this procedure is
given in Annexe 4. During the May workshop interim results of the modelling were shown
along with digitally-altered photographs illustrating how local landscape views could
change (Figure 6.5).

Table 6.5. Land use allocations

Land uses modelled Traditional (n. cells) New Multifunctionality (n.cells)

Agriculture (10%) 275625 (15%) 413438

Coniferous (20%) 551250 (10%) 275625

Broadleaved (10%) 275625 (40%) 1102500

Grazing areas (50%)1378215 (15%) 275625

Eucalyptus (10%) 275625 (20%) 551250

Note: The number of cells was calculated as follows: 1 cell (20*20m)=400 m2=0.04 ha.
Total cells=2 756 250=110 205 ha. A percentage of the total number of cells correspondent
to the description of the scenario (see Table 6.7) was allocated in the land use map for each
storyline.

6.2.4. Scenario evaluation and implementation

6.2.4.1. Workshop

The evaluation of the two storylines was done in two ways. First, participants in the
Gavieira and Entre-Ambos-os-Rios focus groups were told to select the preferred future for
their area of residence as well as to explore further the opportunities, challenges and types
of financial support needed to implement those scenarios. Second, a concise economic
evaluation of both scenarios was made based on DEFRA‟s framework of 1) assessing the
environmental baseline, 2) analyse policy changes effects in the environment 3) value the
173
changes of ecosystems in relation to human wellbeing (DEFRA, 2006). According to this
framework a broad calculation of the costs and benefits was made for each scenario. The
economic values used were based on data from the forestry national institute (DGRF,
2007b). This simplistic analysis intended to stimulate discussion and to be further
improved after consultation with experts in the area. The first evaluation step occurred
during a workshop organised by the lead author in partnership with ARDAL and the
municipal forestry office of Ponte da Barca Municipality whilst the economic valuation
was presented at an expert meeting.

The final workshop aimed to present and evaluate the storylines developed in the
preceding focus group meetings, this time bringing together members of both previous
focus groups. The overarching question was: “What types of institutional arrangements
might be able to deliver these “futures”. To get to this conclusion, three subsidiary
questions were addressed. Those were: 1) who should be involved?; 2) what should be
their role in the implementation of a forest “future”?; and, 3) what type of financial
incentives is (are) required to implement the preferred future?

The workshop started with a presentation of the work developed since January, in which
the participants had been involved. In this initial part of the workshop both groups had
opportunity to share their views. In this plenary session the group discussions were guided
by the following questions: a) are the storylines credible? b) how can these storylines be
improved? c) is there another possible future? d) is there agreement on who will win/lose
in these scenarios?, and, finally e) what is your preferred future? Immediately after the
participants had chosen their preferred future the participants were divided into two
discussion groups based on the “future chosen”.

Working in separate groups in different rooms, each group had its own facilitator who
explained the plan for work as well as the “good practice” rules. Individuals wrote down
their personal opinions about the topics, followed by a group discussion. The topics
addressed were:

1) Opportunities and challenges for the scenario chosen, and,

2) Institutions, financial mechanisms and partnerships able to promote the scenario.

The groups addressed the financial support needed to put the scenario into practice. They
were given tasks such as:

1) Identify sources of income as well as expenses that are likely to be linked with the
implementation of the scenario

174
2) Identify what institution (s) are seen as able to directly manage the preferred future (the
coordinator) as well as those seen as potential partners.

3) Finally, the groups were asked to simulate applying for financial support, by filling in an
application form for a pilot-project. The form to be filled had the following structure: 1)
coordinator of the project, 2) partnerships to put into place, 3) business plan 4) ways in
which coordinator and partners will overcome financial needs- internal financing, 5)
external financial support need, 6) promotion and divulgation of activities, and, 7) steps to
implement the scenario.

6.2.4.2. Implementation

After all the previous phases, it was agreed by the author, the ARDAL and the forestry
municipal office that the results would be better validated by formally engaging the major
forestry institutions in the study area. Accordingly, invitations were made to the regional
office of the major forestry institution (Direccao Geral dos Recursos Florestais), as well as
to the National Park (Parque Nacional da Peneda-Geres). In the invitation letter the
purpose and agenda of the meeting were explained and a summary of the work done was
sent. The forestry institutions had the opportunity to select their team members who were
believed to have contributed significantly during the process of discussions. All the
participants in the focus groups meetings and in the workshop were also invited for the
meeting. The format of the meeting was informal. It involved a presentation of the two
scenarios followed by the presentation of a first draft of the economic analysis of both
scenarios. The discussion was framed by four questions namely: a) are the scenario
storylines credible?; b) How can the economic evaluation be improved?; b1) Are there
economic values for the services provided by forests locally able to improve the analysis
done so far? c) Are these scenarios “good enough” to be tested through a pilot project?

6.3. Results

The case study created, developed and validated two scenario storylines through a series of
participatory processes (two focus groups meetings, one comprehensive workshop and one
expert meeting). One scenario focussed on continuity of the traditional management
patterns, with an emphasis on direct goods such as timber and livestock grazing (traditional
multifunctionality). The other concentrated on indirect ecological services, such as soil and
water protection as well as carbon sequestration (new multifunctionality).

The participants in the focus group meetings for Gavieira and Entre-Ambos-os-Rios were
appointed by different local institutions namely a rural development association- ARDAL

175
and parish council of Entre Ambos-os-Rios. The participants in the two focus groups
reflected the different patterns of usage of their forests. In Entre Ambos-os-Rios the group
had a majority of farmers and forestry landowners as well as forestry related people such
as local fire-fighters. In Gavieira, professional planning and managers as well as farmers
and industry representatives participated (Figure 6.4).

Gavieira Entre Ambos os-Rios

Participants
National Park…
Forestry engineer
Farmer City council
Forestry…
Forestry…
Farmer
Forestry Engineer
Local government
Ardal Fire fighters

0 1 2 3 4 0 2 4

Number
of visits to
forests Once a month Once a month

Once a week Once a week

Twice a week Twice a week

Daily Daily

0 1 2 3 0 1 2 3

Scenario
chosen "Scenario 2 New
"Scenario 2 New
multifunctionality"
multifunctionality"
"Scenario 1 "Scenario 1 Traditional
Traditional… multifunctionality"

"Business as usual" "Business as usual"

0 1 2 3 4 0 2 4 6

Figure 6.4. Professional activity of participants, their usage of forests and scenario chosen

176
Remarkably, in each parish a unanimous decision within the group was reached in order to
select the desired future for each parish. In Entre Ambos-os-Rios the preferred future was
the “traditional multifunctionality” while Gavieira participants clearly preferred the “new
multifunctionality”. These results in Figure 6.4 were possibly influenced by the
provenance and background of the participants. In Entre Ambos-os-Rios the use patterns
are dominated by farmers, hunters and landowners, while in Gavieira participants
represented a mixture of backgrounds and relationships with forests reflecting a more
varied pattern of management.

Yet these two groups, with different backgrounds, had a very similar opinion about the
strengths and weaknesses of the forests in the region. Table 6.6 summarizes the “feelings”
of the two groups about the forests of their region.

Table 6.6. The responses of the stakeholders to the issues raised by multifunctional forest
management in the two case study areas.

PROUD HOPE SADNESS FEAR

“There are still “Better ability to “Depopulation trends- “Forests will

pristine oak forests” manage” Abandonment of disappear due to fire
rural” and abandonment”
“The landscape is “Entrepreneurship
beautiful” from local population” “Litter and scrubs “Bad management”
scattered in forests
“There is a “Policies appropriate “Economic pressures
with its related fire
multifunctional to reverse from new uses such as
ignition potential”
landscape with depopulation of rural carbon sequestration
multiple uses and areas” “Soil erosion” (mainly eucalyptus)
functions that allow a and forestry biomass
“New uses such as “Decrease in primary
multitude of (damage in natural
carbon sink projects sectors activities
livelihoods” regeneration)”
and forestry biomass” (forestry and
“Tradition and culture agriculture)” “Increase in area of
“Arson and bad
shapes the landscape” exotic tree species”
management “Multifunctional land
“ Good policies such punishment” uses that are not viable “Landslides due to
as the one supporting anymore” soil erosion”
local fire-fighters
teams”

Although different in composition, both groups were unanimous in rejecting the base-line
storylines initially proposed, namely, the carbon offsetting and conventional forestry
preliminary scenarios. Instead, both groups emphasised the need to create multifunctional
forests. The opinions of the two groups diverged upon the type of multifunctionality
thought likely to be successful. The next section describes the two contrasting storylines
developed.

177
6.3.1. Description of the storylines
The two types of multifunctional forests named as “traditional” and “new” were seen by
local stakeholders as being able to overcome the unfavourable trends that the forestry
sector has been facing in the region. In Table 6.7 the major differences between the two
types of multifunctional forests are summarised.
Table 6.7. Characteristics of the two multifunctional landscape scenarios
Name TRADITIONAL
Philosophy Continuity: Focuses on
traditional goods such as
timber and non-timber
products though uses a more
efficient approach in order to
increase income.
Type of Integrated
Multifunctionality
Rural Endogenous
Development
Role of forestry Utilitarian/Community
discourse stability
Sustainability Weak sustainability
Economic driver “cultural” tourism. Income
sources: Tourism, timber and
non timber products such as
honey, goat meat, game.
Income generated locally.
Land use ~30 a 40 % forest managed
for timber production and
(in 100 ha)
silvo-pastural as defined in
PROF
~40 % natural grazing areas
~10% agriculture
~10% other uses
Management 60% of the area is
traditionally managed.
Sell products
Protection of Emphasis and ecological care
nature
Institutions able to Local e.g. parish council
promote it technically helped by public
institutions such as National
park, councils and DGRF.
(Bottom-up)
NEW MULTIFUNTIONALITY
Change: Focuses on ecosystem
services and embraces the
challenge of build new markets
for environmental services that
are still in initial stages
Spatial
Exogenous
Nature conservation
Strong sustainability
“active” tourism Income
sources: Tourism, ecosystem
services carbon trade, certified
timber. Income generated
globally
~50 a 60% forest managed for
conservation and protection as
defined in PROF.
~10 % improved grazing areas
~15 % agriculture
~25 % other uses
100 % of the area has
professionalized management
Sell services
Very much emphasis on ecology
and ecological care
High level institution which
manages different interests
(national park, municipalities and
forestry office) (Top-down)
178
One scenario focussed on continuity of the traditional management system well known by
its patchy and mosaic landscape (traditional multifunctionality). The other scenario aimed
at radical changes in both management and the institutional arrangements to deliver it (new
multifunctionality). This was intended to create continuous cover forestry, such as existed
in a pre-industrial age. Figure 6.5 shows the land use maps and the photo-montages that
were used to illustrate the landscape features of each scenario for the two parishes.

In the traditional multifunctionality scenario the major goal was to increase productivity
both for direct goods, namely timber and non-timber products in such a way that the
income generated could attract new rural population into the parish. Because these goods
are the ones traditionally used by the rural communities, the philosophy on which this
scenario was based on is one of continuity.

By contrast, within the new multifunctionality scenario view, emphasis was put on
ecosystem services such as water flow regulation and carbon sequestration. This scenario
also acknowledged the importance of other indirect ecosystem services such as soil
protection and waste assimilation, help to promote the health and vitality of the whole
landscape.

All these measures in the new multifunctionality scenario were seen as requiring
professional management which could create a new form of ecosystem services market that
so far has never been tried in Portugal (an exception is small scale projects for carbon
credits trading by the National Park). These new markets of ecosystem services that forests
provide were though likely to be very important in a climate change context. In this
scenario the removal of the natural capital is minimal, as the scenario is in favour of a
strong sustainability approach. Due to all this, the new multifunctionality scenario
addresses the long term, while the traditional multifunctionality scenario focuses on the
short-medium and more consumist needs (weak sustainability). Because this scenario
assumes a radical change in management, its philosophy is based on radical change.

Another difference between the two scenarios involves livestock grazing. Livestock
grazing and the related income (selling cows and goat meat) is one of the traditional
revenues of rural populations. Recognizing such an issue, in the traditional
multifunctionality scenario, grazing areas form the “matrix” of the landscape (Figure 5-left
side) and livestock circulates “free” within the area. By contrast, in the new
multifunctionality scenario, grazing areas are reduced to small plots where improvements

179
in pasture take place (Figure 6.5–right side) and these will be fenced to avoid damage from
livestock in natural regeneration of oak trees and other broadleaves plantations.

It was envisaged that an integration of functions in the same spatial area would occur in the
traditional scenario with a spatial separation of functions (e.g. pastures from forests) in the
new multifunctionality. As far as rural development discourses are concerned, the
traditional scenario relies on the assumption that goods and services are generated and
provided by the local community without direct interference of external agents
(endogenous). In contrast, the new multifunctionality scenario depends on external markets
(exogenous). In addition, the role of forestry in the traditional scenario matches the
“community stability” discourse, while the nature conservation discourse is prominent on
the new multifunctionality scenario.

There were also differences between the two scenarios relatively to “new” uses such as
forest biomass and carbon sequestration. In the traditional multifunctionality scenario, the
area of those new uses, though significant, was not a prominent feature. On the other hand,
in the new multifunctionality scenario, areas with broadleaved trees managed for carbon
sequestration shape the landscape mosaic.

Another factor to be considered is the way in which the two scenarios influence the
landscape. In the case of the traditional scenario it is known that the mosaic that would be
created is very valued by people (Figure 6.5). The increasing numbers of visitors in the
National Park clearly appreciate such a landscape in which culture and tradition are
embedded. However, at present, this mosaic has not been proven viable. Rural areas are
depopulating and this particular mosaic is not able to reverse such a trend.

On the other hand, the new multifunctionality implies a landscape with big blocks of oaks
and other broadleaves (Figure 6.5) that will resemble native forests in Europe (centuries
ago), but those forests were largely destroyed during subsequent “development” processes.
Another issue is the size of the area that needs to be afforested in order to implement the
new multifunctionality scenario. In addition to the cost of reforestation, care needs to be
taken to minimize the impacts that the new plantations could have on the scenic beauty of
the landscape in the only National Park of Portugal. According to work in Chapter 5, the
effect of the plantations in the landscape could be minimized if the shape of the new
plantations will be irregular with uneven stands of a mixture of tree species. It seems that
these are the characteristics which people in the region most value (Chapter5).

180
 TRADITIONAL NEW MULTIFUNCTIONALITY
MULTIFUNCTIONALITY

Land use
models

Gavieira

Entre-
Ambos-
os-Rios

Figure 6.5. Multifunctional scenarios as developed and presented in the case studies
Larger format photos are shown in Annexe 5.

6.3.2. Scenario evaluation and implementation

An overarching result of the workshop was an agreement amongst the group of participants
that the current situation will not be able to deliver a sustainable landscape. Thus there is a
181
need to make changes in institutional arrangements, policies and review financial
incentives in order to achieve multi functional forests in Minho region of Portugal. Another
outcome was that the storylines of the two scenarios (traditional and new
multifunctionality) were accepted by the stakeholders. Even more remarkable is that in
each parish a unanimous decision within the group was reached in order to select the
desired future for the parish. According to the group these two possible futures represent
two different strategies, both of which could deliver a viable multifunctionality in the local
forest landscape.

For each scenario the participants were able to list and evaluate the various opportunities,
challenges and institutions needed to put each scenario into practice (see Table 6.9). The
participants who had chosen the traditional scenario referred to its opportunities regard to
the appropriateness of the scenario for a protected area such as the Peneda-Geres National
Park. Other opportunity highlighted was the familiarity that local communities have
already with this type of management approach. The new multifunctionality scenario deals
with uncertainties by safeguarding critical natural resources, thereby maintaining more
resilient environmental conditions. This can be very valuable when all manner of
environmental, economic and social uncertainties are faced. However, successful
implementation is only achievable through engagement of a wide range of institutions and
local stakeholders. This scenario also requires sharing of responsibilities between public
and private landowners. Because it generates indirect services which are difficult to
perceive by the general public, institutions have a major role in order to “make obvious” to
the stakeholders the benefits that this scenario is likely to generate. Thus it is very
important to have a public-private partnership in which institutions engage communities in
this process of change in ecosystem management. If that does not happen, the new
multifunctionality scenario and its ideas, will only be an “innovative academic thought”.

In the traditional multifunctionality scenario major challenges identified by the

stakeholders were: 1) improving the relationship between local communities and the
National Park; 2) defeating bureaucracy in order to bring investments namely for tourism
related activities, 3) transcending property boundaries and achieving the whole landscape
management. Challenges for the new multifunctionality scenario were related to: 1)
regulation of uses namely by restricting the areas for livestock grazing; 2) overcoming
problems of fragmented landownership and transcend property boundaries; 3) bringing
entrepreneurship and active populations to rural areas.

According to participants the two scenario storylines will require two completely distinct
institutuional arrangements. While the traditional multifunctionality scenario requires a
182
local based “governance”, namely by a partnership of parish councils with local NGOs
aided by the technical expertise of the National Park, the new multifunctionality scenario
needs a higher level institution with professional managers able to deal with incipient
markets and finding other financial means for cooperation between upstream and
downstream municipalities in order to promote whole catchment management (Chapter 4).

Table 6.8. Landscape management implications arising from the comprehensive workshop

TRADITIONAL NEW MULTI FUNCTIONALITY

“More appropriate for the type of landscape of “Explores the protection throughout economic
Peneda-Geres National Park (cultural valuation of ecosystem services such as soil and
landscape)” water protection”

“Local communities are already familiar with “Will allow a “viable planning” of a mixture of
the assumptions of this scenario” “Generates blocks within the landscape such as pristine areas,
Opportunities and challenges

income that goes directly to the farmers and this improved grazing areas, sustainable agriculture,
will increase “attractiveness” of rural areas” sustainable forestry ...”

Most suited to local culture and economy yet
enhancing ecological integrity.
“Revert depopulation trends”
“Will be able to integrate traditional and old uses
throughout technical expertise in order to
minimize conflicts between uses (eg.fenced
grazing areas to not damage new forests)”

Improve the relationship between local
population with Peneda-Geres National park
“lack of dialogue grows tension”
“Establishing new rules for users”
“Need to engage local stakeholders into this
process of change”

“Creation of tourism infra-structures that do not
exist at present”
Defeat bureaucracy- “local farmers need to fill
in too many forms”
“transcend property boundaries”
“Need to show economic viability and well
defined institutional arrangement to gain
credibility” “Overcome problems of fragmented
land ownership” “transcend property boundaries”
“Bring new entrepreneur and active people to the
area”

“Local parish councils backed up by “Need to create a more professionalized institution

professionals” that will be able to do integrated management of
the whole landscape”

Geres National Park (preferably with a “Forests can‟t be the focus rather different
Institutions

cooperative approach towards local populations interests groups (tourism, agriculture...) should
via the municipalities) work together- eg create a trust or a foundation to
manage the whole mountain”

“Municipalities in league with national park and “Two major goals are professionalism and
other planning bodies” integrated management of all the resources”

“Local development organizations”(eg. ARDAL “The first task of this Foundation is to engage with
and ADERE) local stakeholders showing the economic benefits
likely to occur”

The last stage of this study was based on an “expert-meeting session” on which the ideas
developed throughout the focus groups and the workshop were shown to the two major
institutions namely Direccao Geral dos Recursos Florestais (DGDR) and the Peneda-
183
Geres National Park. Overall the experts considered that both storylines were viable
though National Park officials clearly favoured the new multifunctionality scenario while
the national forestry office was more sympathetic with the traditional multifunctionality
scenario. Although guided by a series of questions (see Section 6.2.4.2) the discussion was
long and sometimes confusing. The economic evaluation was barely commented upon and
none of the institutions offered any further indication of other possible sources of
economic data. The experts raised aspects to possibly overcome the depopulation trends. In
their opinion, both scenarios need to tackle that matter first, because without people with
viable livelihoods, none of the scenarios are likely to be socially and economically
credible. Another issue raised was how to sustain livelihoods and conserve nature in an
area with fragile ecosystems prone to erosion. One of the topics addressed was the
relevance of small scale projects using forestry biomass for heating buildings which were
seen as an opportunity to overcome the depopulation trends. It was also said, however, that
this matter has been approached instead by supporting large scale infrastructure projects
that are planned to be located in nearby urban centres.

The National Park was not supportive of the implementation of the traditional
multifunctionality scenario within the national park boundaries. As a result a second
version of the traditional multifunctionality scenario was made this time increasing the area
of broadleaves trees and diminishing the area of coniferous trees such as maritime pine.
Another outcome of this meeting was that there are planning instruments, namely the
“Zonas de Intervencao Florestal” (ZIFs) that can be used to implement both scenarios. A
ZIF creates a joint management plan for forests belonging to a group of forest landowners.
The overall goal is to overcome the fragmented land tenure system by giving economic
incentives for communal forest management. There are also legal frameworks on which
the management strategies can be supported by legislation into place such as PROFS.

Despite the support for the two storylines, there was no consensus amongst the expert
group concerning any viable means to test the implementation of the scenarios. The
National Park, with the consent of the research team, introduced the two storylines in its
management plan, which was under revision at the time the study was developed. Despite
this, neither the National Park nor the regional forestry office fully supported the
implementation of the storylines, recognising difficulties in building cohesive governance
that would promote their implementation via a pilot scheme.

184
6.4. Discussion

Managing forests in a way that user groups, sustainability practitioners and forestry
institutions agree upon is not easy (O' Riordan and Stoll-Kleemann, 2002; Sheppard and
Harshaw, 2001). For agreement to occur, via viable landscape design procedures, through
which multiple functions (production, environmental protection and recreation) may be
coordinated by means of innovative planning, there is a need to negotiate a set of
commonly agreed objectives and shared responsibilities (Milligan and O'Riordan, 2007).

In this study, although “viable future” storylines were created, validated and supported
from local stakeholders according to the state of the art participatory approaches (Milligan
and O'Riordan, 2007) and stakeholder engagement techniques (Grimble and Chan, 1995),
the workshop and expert meeting revealed there were neither robust planning mechanisms
nor adaptive governance systems with the capacity to put into place “futures” likely to
deliver more sustainable landscape-scale uses in these areas. Consequentely, the research
illustrates the difficulties in forging governance systems that have the capacity and the
vision to be able to put the sustainable development concept into practice, in a policy
setting that is confused, contradictory, and where the “status quo” tends to be given
prominence (Milligan and O'Riordan, 2007; Nicholson-Cole and O'Riordan, 2009). This
study therefore suggests that both the conceptual approach to sustainable forestry
management, and the present institutional arrangements and outlooks, are not compatible
for dealing with whole landscapes on a multifunctional basis. In two sides of the same river
this study found that different communities are likely to see different “futures” for forestry
in their area of residence corroborating the idea of a much diverse rural Europe (Elands
and Wiersum, 2001; Pinto-Correia and Breman, 2008). This conclusion challenges the
current planning and governance systems in place, thus compromising the implementation
of SFM in Portugal.

There is, as yet, no formal planning framework in Portugal for whole landscape sustainable
management. Nor is there a property rights basis for engaging landowners to cooperate on
a fully collaborative arrangement. While there are planning and management rules in
Portugal for approaching these matters, e.g. the creation of a ZIF in each parish (Table
6.1), they are not strong enough, nor are they sufficiently institutionally and financially
supported, to work on a coherent multifunctional basis. The legal framework for the ZIFs
is under recurrent re arrangement. With such an inconstant institutional and financing
system it is difficult to successfully implement long term revenues in a sector such as
forestry.

185
This research has provided a basis for moving forward. The method of establishing whole
landscape preferences, set in a framework of user values and expectations, and
geographically embedded and landscape cultures and histories, provides a basis for
mapping and revealing multifunctional whole landscape scenarios that are both well
grounded and comprehensible to a wide range of stakeholders (Appleton and Lovett, 2003;
Milligan and O'Riordan, 2007). Another novel contribution of the research was the
establishment of relationships between rural development discourses (Elands and
Wiersum, 2001) with the type of multifunctional land uses (integrated and spatial) that they
are likely to create (Brandt et al., 2000). However, the test of its implementation was
hampered by the weakness of governance systems currently in place.

There is still a dysfunction between the viewpoints of conservationists and National Park
planners who appear to seek to reintroduce indigenous biodiversity, and those at the
municipal level of local politics, who prefer to promote mosaics of land use and economic
diversity. There are still unsolved issues related to the reconciliation of sustaining living
rural communities and ecological integrity of areas such as National Parks.

If there is a call for approaches less focussed on subsidies and more grounded in
investments (OECD, 2006) there are not yet mechanisms able to develop the path for
investing in forestry for small scale heating projects. In the two storylines forestry biomass
was seen as a way to move forward, but the way that Portuguese government is dealing
with this issue by supporting large scale projects reduces the opportunities of rural areas to
create small scale heating projects that will likely contribute to overcoming depopulation
trends as has being demonstrated viable for other parts of rural Europe (Okkonen, 2008).

Therefore, the basis for moving forward is to coordinate the machinery of planning,
financing and management through four levels, namely the EU though its environmental
and regional funds, the Portuguese nation through its regional development mechanisms,
the sub-regions of Portugal through their rural diversity arrangements, and finally, to the
municipalities at the local level. This can best be done by exploring emerging EU policy of
rural revival and the advancement of viable rural economies, with national policies for
sustaining local land use and local communities.

The two multifunctional landscape scenarios offer a way to approach this requirement for
reconciliation and consensus building. The role of GIS maps, rooted in stakeholder
understanding and aspirations, supported by the overlaying photograph images, forms a
basis for establishing this dialogue (Appleton and Lovett, 2003).

186
There is perhaps a case for a specially convened policy/governance workshop, arising from
this research, through which all the relevant parties can meet and debate the creation of
additional planning and financial arrangements for multifunctional sustainable forest
management in Portugal. For this workshop to be truly successful, a broader group of
stakeholders (e.g. including tourism and energy sectors) needs to be engaged.

6.5. Conclusion

“New paradigms” for sustainable rural development have been suggested namely by
OECD (2006) calling for approaches that focus on places instead of sectors (e.g.
agriculture). The research presented here falls within this new paradigm even though, there
were no governance systems willing to take up the challenge to test the implementation of
the “desirable” futures as seen by local stakeholders. There is much to play for, since the
current debate in Europe on viable rural futures is coinciding with a fresh look at the
commercial, economic, and social advantages of living with and for nature. Such a
perspective has to return to life as Europe struggles for its soul and the aftermath of
profound economic uncertainty, huge social distress and well intentioned, but ultimately
ineffective political hand–wringing.

187
“If human understanding of nature is imperfect, then human interaction with nature should
be experimental. That is, policies should be designed and implemented as experiments
probing the behaviour of the natural system. Experiments often surprise and scientists
learn from surprises. So if resource management is thought of from the outset as an
experiment, surprises are opportunities to learn rather than failures to predict”

Lee, N.K. (1995:225)

CHAPTER 7. CONCLUSIONS: The role of multifunctional

forests in sustainable landscapes

188
This concluding chapter is framed by the set of questions originally posed in Chapter 1
(Box 1.4). It provides a synthesis of the research presented in Chapters 3 to 6 which
addressed the particular questions at specific scales presented in Figure 2.7 of Chapter 2.
Sections 7.1 to 7.3 explain how the original questions were addressed. Section 7.4.
describes difficulties encountered in this research, caveats in the use of its findings, and
explores directions for further research. A concluding statement is given in Section 7.5.

7.1. How can sustainable landscape planning and landscape governance

guide forest planning and management?
The goal of this research, as defined in the introductory chapter of the thesis, was to
explore some of the guiding principles for implementing sustainable forestry management
(SFM) at different spatial scales in such a way that forests might contribute to more
sustainable landscapes. It specifically addressed the ways in which planning and
governance concepts, tools and methods might be used to inform SFM at a range of scales
in Portugal.

Landscape planning and landscape governance concepts, tools and methods were crucial in
providing a framework for guiding forest planning and management geared to sustainable
goals in Portugal. Planning tools such as landscape metrics allowed the study of forest
composition (PS, PLAND) and spatial arrangement (PD) which helped to inform place-
specific strategies to improve forest management (Chapter 3). By incorporating public uses
and preferences for forests into multifunctional plans, it was shown that there is a need to
vary the type of forestry across the area of two watersheds in the Minho region (Pinto-
Correia and Breman, 2008). This can be achieved by creating different
“multifunctionalities” illustrating ways in which some forest functions are better provided
if integrated (integrated multifunctionality) or separated in space (spatial multifunctionality
or dominant use) across the area of the two river basins (Chapter 4) (Mander et al., 2007).
The analysis of public preferences for forests (Chapter 5) showed that different
stakeholders groups have contrasting preferences for forest characteristics. These findings
are in accordance with the findings of others (Harshaw and Tindall, 2005; Misgav, 2000).
The analysis of public preferences for forests, using two criteria (attractiveness and
management), revealed that not only scenic beauty, but also what is perceived by
stakeholders as “good management”, might undermine the ecology of forests ecosystems
(Gobster et al., 2007). This raises concerns about important ecological functions such as
soil and water protection (Chapter 5) that, as shown in Chapter 4, are important in forging
strategies able to reinforce development across urban and rural regions (CEMAT, 2007).
Finally, in developing scenarios for sustainable forestry management (SFM) it was shown

189
that the ways in which forests might be integrated in the whole landscape are likely to vary
in different parishes with similar socio-economic conditions within a single watershed. It
was also pointed out that there is a need for a new governance approach to implement
successful planning (Chapter 6), as there were no governance systems in place to test the
implementation of any of the scenario storylines.

The tools, concepts and methods used in this work (Table 7.1) were helpful to guide forest
management at different spatial scales. This set of tools might be used by 1) decision
makers, 2) planners and 3) forestry practitioners dealing with the implementation of
sustainable forestry management, as follows

1. Decision-making (at the local scale): parish councils are often confronted with decisions
about alternative future investments. Promoting a set of discussions, framed by questions
such as those presented in Table 6.3 in Chapter 6, may help to better address alternative
futures and help in developing participatory processes for decision-making. Furthermore,
the set of questions proposed are likely to facilitate the setting of priorities and decisions to
be made. Using visual images such as photographs proved to be very effective in engaging
the public for the study of public preferences for environmental conditions. Moreover, as
was shown in Chapter 5, it is important to create awareness amongst the public that there
are different criteria for forest management and that public preferences for forests might
threaten some “invisible” ecological functions. This calls for partnerships between
researchers, forestry practitioners and decision makers (Section 7.3.3).

2. Guiding the planners: Planners are frequently asked to provide “remedies” that can
guide favourable development patterns in an appropriate timing at a reasonable cost. The
set of landscape metrics presented in this thesis might help to set possible guidelines for
more detailed planning actions. For example, by using a set of class metrics that are
relatively easy to interpret, such as those used in Chapter 3 (NP, PD and PLAND), it is
inexpensive and straightforward to obtain a broad picture of the environmental conditions
in an area. Furthermore, Chapters 3 to 5 of this thesis focused on specific aspects of
planning that allowed the study of: i) forest structure by focussing on the analysis of forest
landscape metrics across different socio-economic regions (Chapter 3), ii) forest
functioning, by suggesting different forest management strategies for contrasting socio-
economic and environmental places (Chapter 4) and iii) landscape evaluation, in which
public preferences for forests were surveyed (Chapter 5). These might be seen as different
planning stages as defined by Botequilha Leitao and Ahern (2002) and the approach
presented here touches on the different planning phases.

190
3. Guidance for practitioners: For practitioners this toolbox might help to give context for
afforestation projects implemented at a local scale. For example, forestry projects for
afforestation must take account of public preferences for forests, providing a mixture of
tree species as well as uneven stand structures (Chapter 5). However practitioners need
also to be aware of the implications of what is understood as good practice. For example,
fire management might reduce the forests‟ ability to provide other services such as soil and
water protection (Chapter 5). Foresters need to be more aware that afforestations projects
on the ground are often threatened by external factors arising from neighbouring land uses.
Hence, placing forests within the whole landscape has to be one priority for forestry in the
21st century. The toolbox used throughout this work was able to address forestry within the
whole landscape mosaic, as illustrated in Table 7.1 below.

The multi scale approach adopted was useful in the sense that what was depicted at larger
spatial scales was further refined when smaller spatial scales were addressed (Herrmann
and Osinski, 1999). For example, at the national scale no association was found between
the characteristics of the forests and the level of socio-economic development. On the
contrary, when the analysis focused on the Northern region, landscape metrics were used
revealing that at this scale an inverse correlation between higher levels of socio-economic
and better forest condition appeared (Chapter 3). The work at regional scale also suggested
that different strategies are likely to be needed in order to move towards SD.

Going even further down the scale of analysis, this time focussing on two watershed
catchments located within the regional scale of analysis, it was possible to better address
how management strategies are likely to differ across the gradient depicted at the higher
scale of analysis. When analysing landscape metrics in the Northern region of Portugal it
was suggested that different strategies would be needed to implement forest management
in urban and rural regions (Chapter 3). The research at the watershed scale further
demonstrated the ways in which strategies can be put into practice by varying the type of
multifunctionality (integrated vs. spatial or even dominant use). It also revealed that there
were differences in uses of forests across the rural/urban gradient. Forests in rural areas
were used for a multitude of goods, while in urban areas the major uses were production
and recreation (Elands and Wiersum, 2001). It was also evident that forest conditions are
far from demonstrating sustainable characteristics. Therefore, forest management practices
aiming at, for example, increasing the age/diameter classes of forests in the study area are
of utmost importance (Nabuurs et al., 2001; Pommerening and Murphy, 2004). Still at the
watershed scale, the study found variations in public preferences regarding forest cover
and stand age across different user groups (Chapter 5), this finding being in accordance
191
with a large body of literature on public preferences for natural resources (Abello and
Bernaldez, 1986; Harshaw and Tindall, 2005).

Table 7.1. The toolbox used

Scale Goal Tools and methods

National Study how forests 1) Group regions according to socio-economic

vary in contrasting characteristics (e.g. PCA and CA)
/Regional
regions in order to
2) Calculate simple class metrics (or landscape
set appropriate
metrics)
management
strategies 3) Correlate socio-economic conditions of groups
of municipalities with the characteristics of their
forests (metrics)

River basin Study people‟s 1) Field survey for evaluate forest condition
preferences
2) Questionnaire survey: uses, public preferences
3) Evaluate “desirable” multifunctionalities ( the
distribution of forestry functions across the area)

River basin Study people 1) Survey uses of forests and group stakeholders
preferences for accordingly
forests
2) Photo-questionnaire: showing forests with
different characteristics
3) Study public preferences for forests

Parish Study of viable 1) Focus group meetings for develop scenario

futures for allocating storylines.
different land use
2) Land use modelling for allocate land uses
types
according to storylines developed
3) Present the outcome of the modelling using
visual tools

The smallest scale analysed in the research was the parish scale. At this scale, an in depth
study was conducted to investigate the type of forest that stakeholders see as viable for the
future. As this was a relatively small area, it was possible to define the type of
multifunctionality that stakeholders see as viable for two parishes. As such, this piece of
work further refined the achievements of broader spatial scales. Chapter 4 concluded that
multifunctional forest management was of utmost importance and Chapter 6 further
demonstrated the type of multifunctionality that stakeholders see as desirable. Table 7.2
shows the different research stages.

192
 Table 7.2. Research stages

2005 2006 2007 2008

(National/ (River basin) (Parish)
regional)

April- January, March May July

Beginning of
PhD

Gather data, Field survey Questionnaire Focus group 1 and Workshop Expert
build socio- (photos were survey (including 2 (in each parish) meeting
(Two groups
economic and taken) the photo-
(creating and together as well as (Local and
forests Analyse of questionnaire)
developing national park regional
metrics of socio-
storylines) representative) forestry
databases economic and
institutions)
forest metrics
data

One parish aimed at forestry to sustain living communities, while the other preferred to
implement forest management for nature conservation (Chapter 6). The findings of this
study are illustrative of the diversity of rural areas in Europe (Elands and Wiersum, 2001;
Nabuurs et al., 2001; Niskanen and Lin, 2001; Slee, 2007a), highlighting that forests may
have different role(s) across landscapes. Even within a rural area with similar
characteristics (Chapter 6), two contrasting futures were envisioned by different
stakeholder groups. These findings are in agreement with a substantial body of literature
reporting that there is no panacea in management of natural resources (Ostrom, 2007;
Ostrom et al., 2007). It also shows that if the two local scenarios were implemented this
would be likely to create a multifunctional landscape able to fulfil the preferences for
forests of a broader group of stakeholders. It was reported in Chapter 5 that there are
people (single users) who preferred to see continuous cover forestry (Photo FC), such as
the new multifunctionality scenario would be likely to deliver. Meanwhile other
stakeholders (multi-users) valued a mosaic landscape as shown in Photo FA. Therefore, the
combination of the two scenarios at watershed scale would be ideal and in line with the
wishes of a broader group of stakeholders.

The two scenarios were different in many aspects. One scenario clearly implied the notion
of endogenous (traditional multifunctionality) development, while the other was based in
the exogenous (new multifunctionality) model (Okkonen, 2008). Similarly, in the
traditional multifunctionality scenario the notion of “weak sustainability” was implicit in
the sense that it was short-term-driven. In contrast, the new multifunctionality implied
“strong sustainability” in which nature conservation would be the overarching goal
(Costanza and Daly, 1992; Costanza et al., 1997). When looking at the two storylines,
193
based on a temporal perspective, it can be suggested that the traditional scenario might be
transformed into the new multifunctionality scenario in the longer term. That is to say, if
environmental services such as water and soil protection were to become the focus for
whole catchment management then it is possible that the traditional multifunctionality
scenario would evolve towards the premise of the new multifunctionality scenario. At
present, stakeholders preferring the traditional scenarios are not able to understand the
ways in which partnerships for whole catchment management can be put into practice. As
such, they prefer a scenario that they are able to understand. Furthermore, environmental
services provision and whole catchment management is, so far, still in its infancy.

One issue that this research does not fully explore (although the implications of the
scenario storyline at landscape level were described) is the ways in which these two
scenarios may be geographically compatible (Hersperger, 2006; Hersperger and Forman,
2003). It is likely, however, that the combination of both scenarios within the area of the
watershed will deliver a diverse landscape mosaic that will create ecosystems able to
support a variety of animal and plant species. Furthermore, as the new multifunctionality
scenario is driven by external demands (increasingly affected by external factors e.g.
carbon markets) which are not controlled locally it is more susceptible to international
drivers. On the contrary, the traditional scenario is more focused and dependent on local
strategies. Also by these reasons the combination of both at the landscape level seems
ideal.

The framework of tools, concepts and methods employed in this thesis began by using
approaches of quantitative character, at larger spatial scales (landscape metrics,
questionnaire survey). Qualitative approaches (e.g. focus group meetings) were later used
at smaller scales (parish) (Flick, 2002). By conducting the research in this manner, the
work developed at smaller scales was framed by the achievements of the higher scales of
analysis. This allowed effectively use of the time requested of stakeholders, avoiding time-
consuming processes that might cause participatory “fatigue” (Flick, 2002). Developed in
such a manner, the work of the scenario creation and development took four days of
stakeholders‟ time (one day in each focus group, one day in the workshop and one day in
the expert meeting). Although this approach was justified because it goes from the general
to the specific (always setting the frame for the smaller scales), it also had its drawbacks
(Flick, 2002). It is recognised that quantitative approaches are less able to deliver a
complete understanding of complicated socio-ecological systems (SESs) thus overall
trends might be misunderstood (Antunes et al., 2006). For example, in the case of the
questionnaire survey it was apparent that forestry in the future is likely to be influenced by
194
new uses such as carbon sequestration and forestry biomass, in opposition to more
traditional uses such as grazing. With the results of the questionnaire in mind two baseline
storylines for scenario development were proposed (conventional forestry and green
forestry) and presented in the initial focus group meeting (Chapter 6). These preliminary
storylines were, according to the stakeholders‟ views, completely inappropriate to the area
of those parishes.

If, instead of starting the study by addressing broader scales using quantitative approaches,
the work had focused initially on the analysis of smaller scales (in which qualitative
approaches are more effective) a more robust approach based on a better understanding of
the issues at stake might have been implemented. (Antunes et al., 2006; Flick, 2002).
Despite recognising the effect that the approach used might have on the results of this
research, it is fair to say that the approach used (first quantitative and qualitative later) was
effective in the sense that it allowed for more effective time management. If qualitative
approaches had been employed in the beginning, it is likely, that the research would have
required more time for completion (Flick, 2002) due to researcher inexperience and this
could have compromised the analysis at other spatial scales.

Summarising, sustainable landscape planning concepts, tools and methods contributed to

the study of the role of forests in different settings. This holistic approach is able to
broaden the goals of the forester by including forestry within the whole landscape (Dolman
et al., 2001). If it is increasingly recognised that forests gain from being studied at
landscape scale, modern forestry has to i) find viable multifunctionalites for forests across
different settings (Pinto-Correia and Breman, 2008), looking for opportunities to interact
with other sectors such as tourism and energy (Dwyer, 2007) (Chapter 4) ii) engage the
public in forest management (Chapters 5 and 6), in order to explore the varied functions of
forests across different socio-economic settings, not only based on forest scenic beauty
(Daniel, 2001), but as a wider development opportunity (Dwyer, 2007) and finally iii) be
capable of coordinating the implementation of SFM across different scales.

7.2. What roles might multifunctional forests have in sustainable

landscapes?
Forests may contribute to sustainable landscapes in a range of ways, by providing different
combinations of goods and services such as timber and non-timber products, biodiversity,
water and soil protection, recreation and carbon sequestration to cite a few examples (Slee
et al., 2004; Slee and Snowdon, 1999; Spiecker, 2003; Stengera et al., 2009). Chapter 4
illustrated the ways in which different ecological functions of forests are able to deliver
services across the area of two watershed catchments in Northern Portugal. According to
195
stakeholder opinions the services provided by forests are very important in sustaining the
landscapes of the region.

In order to provide appropriate goods and services, the findings of this study suggest that
planning and governance systems in the forestry sector need to be better coordinated across
scales. As this study has demonstrated, there are not, at present, mechanisms in place that
will allow for the implementation of futures that forest stakeholders see as desirable. At the
local scale, one of the issues that requires further attention is the discrepancy between the
viewpoints of conservationists (National Park), who seek to reintroduce indigenous
biodiversity, and those in local level politics who prefer to promote mosaics of land use
and economic diversity. In addition to these different visions, what, in my opinion,
ultimately undermined the aim of testing the implementation of the scenario storylines
through a pilot scheme was that at higher hierarchical scales mechanisms able to enforce
local scale dynamics were absent. Coordinating bottom up and top down approaches is
crucial in implementing sustainable forestry management in Northern Portugal. Figure 7.1
suggest approaches that might contribute to this coordination.

In order to test the implementation of both local scale scenarios there was a need for the
following coordination at the catchment scale: 1) a leader (or several) to embrace the
challenge of starting the scenario implementation and 2) creation of partnerships amongst
stakeholders in order to share responsibility for its implementation. Landscape is the scale
at which conflicting uses need to be coordinated and compatibilities between type of
forestry pursued (Tress et al., 2001). As such, it is suggested that the implementation of the
two scenarios needs to be put into practice at the catchment scale. In the Lima watershed,
where the local scenarios were created and developed, there were no partnerships created,
nor was there a leader. There was also an absence of mechanisms in place, for example at a
higher scale (the Northern region), that would allow the implementation of either type of
forestry and thus, even if leaders could have been found, their role was likely to be
compromised due to non-existent measures to help accomplish the goal.

196
 EUROPE

Strenghetening of environmental
C
policies
R Transcend
property
O
PORTUGAL boundaries
S
Integration of forestry sector Urban-rural
S
within formal planning partnerships

S Restore
confidence
C in
NORTH A institutions
L T
Creation of a “portfolio of Sectoral
E integration E
measures for different types of
forestry (agriculture, S
forestry,
C T
energy)
CATCHMENT O
Enhance
O
Encourage partnerships, leadership mechanisms
to implement, address functional R to public-
compatibilities between different
D private
types of forestry
I investments

N Engage
LOCAL public into
A
(PARISH) forest
T
management
Identify the type of forestry (ies) I
stakeholders‟ see as viable. O
Participatory tools. The role of
science, transdisciplinarity. N

Figure 7.1. Cross scale coordination to implement SFM at the landscape scale

197
Because forests characteristics within Portugal vary from the North to the South of the
country (Chapter 3), I would be in favour of coordinating the different roles of forests at
the level of the Northern region. At this level, a different portfolio of policies sufficient to
promote different functions needs to be made available. For example, in the cases in which
local communities are supportive of a traditional approach, a combination of measures able
to promote the certification of products such as meat from cows and goats as well the
products derived from them (e.g. cheese, handicrafts made with wool), linked with cultural
activities such as transhumance9 has to be put into practice as packages (Figure 7.2). On
the other hand, there are cases in which local communities value the new
multifunctionality aspect of forestry, measures such enforcing the linkage between rural
areas and urban centres through the provision of environmental services such as soil and
water protection. For this to happen, services provided by forests in the top of the
mountains have to be the subject of ecosystem valuation and partnerships between rural
and urban municipalities by means of the Water Framework Directive need to be supported
to enable whole catchment management. This could create and put into practice ecosystem
goods and services partnerships across regions as is already happening in the US and
France (Johnson et al., 2002). Both the traditional and the new multifunctionality types of
forestry assume that agriculture is a “companion” of forests. Consequently, schemes such
as environmentally friendly farming will be important in both cases.

Traditional multifunctionaity New multifunctionality

Small scale Flood

House management
forestry Environmental
heating
biomass farming schemes
Urban-rural
Cheese partnerships
Cultural Biodiversity
traditions Goat and Carbon Environmental
(folklore) sheep meet market services

Figure 7.2. Portfolio of measures for different types of forestry

For this portfolio of measures to be available in the Northern region of Portugal it seems
essential that forest planning and management in Portugal is better integrated with the
tourism and energy sectors. In addition, it will be of value if EU policies become more

9
Transhumance is the seasonal movement of people with their livestock over relatively short distances,
typically to higher pastures in summer and to lower valleys in winter
198
integrated. So far, measures promoted by institutions such as the EU namely the carbon
markets and Natura 2000, influence the forestry sector but there is no coordination
amongst the different measures and some of them may be spatially contradictory. For
example, with the goal of sequestering carbon, forests may be afforested with eucalyptus
trees, as it is known that eucalyptus trees do efficiently sequester carbon. However, this
may undermine the biodiversity measures that are also the focus of EU policies. In
forestry policies at the EU level, Member States have sovereignty over implementation. In
countries such as Portugal in which forests are an important landscape feature and where
there are difficulties in implementation and delivery of policy mechanisms, I would be in
favour of more effective guidance from the EU. The EU rural development “package”
needs to be coordinated across the “machinery of planning and management” through four
levels. Namely the EU by strengthening the links between its environmental and socio-
economic policies, the Portuguese nation through its regional development mechanisms,
the sub regions of Portugal through their rural diversity arrangements, and finally to the
municipalities and parishes at local level (Figure 7.1).

As shown in Figure 7.1, this coordination should create conditions to test the
implementation of storylines such as those developed in Chapter 6 of this thesis. This test
is crucial in order to better understand the dynamics between ecosystems and society as
pointed out in the body of literature on Socio Ecological Systems (SESs) (Folke et al.,
2002; Folke et al., 2005; Lee, 1995). Critical aspects that should be tackled both from
bottom-up and top-down perspectives include the transcendence of property boundaries,
promotion of urban-rural partnerships, restoration of institutional confidence and
engagement of the public in forest management (Figure 7.1). For this to happen there is a
need for multi level governance systems.

In recent years, sustainable development has been introduced at all levels of Portuguese
planning and governance systems from national to local scale. Changes have been made by
bodies such as the Portuguese Commission for Sustainable Development and the Ministry
of Agriculture and the Environment in order to better manage forests and support
integrated approaches for sustainable forestry management. However, planning and
governance for sustainability in Portugal has yet to become more integrated between
sectors, in order to move towards more sustainable landscapes (Fidelis and Sumares,
2008).

Formal planning in Portugal acknowledges that the ecological balance of landscapes is of

utmost importance. It is evident from the regulatory bodies in place that enormous efforts
have been made, as far as planning regulations are concerned, in order to “steer” away
199
from unfavourable patterns of development (Chapter 2). Despite such efforts, the lack of
coordination on the ground reduces the prospects for the creation of a post-modern attitude
towards the Portuguese landscape (Fidelis and Sumares, 2008). Some of the reasons for
this are: 1) it is not through a sectoral effort that a “whole landscape approach” can be
achieved 2) more than focussing on elaborated sectoral planning, it is necessary to
integrate sectoral interests in order to govern for sustainability, 3) it is necessary to go
beyond planning, and focus on governance systems able to put successful planning into
practice. Governance for sustainability will require effective partnerships between public
bodies and the private sector that are not yet in place (Chapter 6). Furthermore, there is a
need to transcend property boundaries as stated by the groups of stakeholders involved in
the scenario creation and development (Table 6.8, Chapter 6).

In England, the Natural England agency, created in 2006, is seen as one step closer to the
integration of different functions (Southern, 2008). In Portugal an institution such as
Natural England (Natural Portugal?) will, in theory, be of value for integrating forestry
with agri-environmental measures. However, as discussed in Chapter 2, the agencies
created in Portugal so far have not been able to deliver efficiency and this may undermine
the purpose of creating a new organisation (Araújo, 2001). In my opinion, reforming
current institutions at a regional scale, and giving them the goal of integrating functions,
will be more successful than creating a new agency.

Although the existing legal framework holds potential for the implementation of sensitive
landscape planning based on the integration of ecological principles and economic
development, successes of implementation are more likely to be achieved through
integrated planning and governance approaches. As far as governance is concerned, one of
the measures proposed by Andersen and Castelbranco (1993), namely to include an index
based on area of ecological reserve (which can work as an incentive to the municipalities
to protect nature and their cultural heritage), is likely to contribute to better integrated
ecological principles and economic development. Forests are not yet integrated in formal
planning nor are there any public-private partnerships for forest management. This
undermines the potential role(s) of forests in more sustainable landscapes in Northern
Portugal. For this to happen, the forestry sector has to be seen not in isolation but
integrated in the tourism and energy sectors. This will allow the creation of a portfolio of
measures for particular regions with specific forestry vocations (Figure 7.2).

If this multi-scale and cross sectoral coordination does not happen (Figure 7.1), forests will
continue to be neglected and this will accentuate the well known issues, such as wild forest
fires, which threaten the functionality of the entire landscapes. Wild forest fires already,
200
unfortunately, demonstrate that forest management requires attention from society,
researchers and governments alike. It is unfortunate, though likely, that only when a major
catastrophe occurs, will forest management become a priority in Portugal.

If full integration of sectors such as tourism, agriculture, transport and energy are put into
practice, as well as coordination of governance strategies across scales, forests in the
Northern region of Portugal are likely to have all the five roles described by Elands and
Wiersun (2001). In some areas in which agriculture still flourishes forests might be
regarded as an important activity to complement agriculture (agri-ruralist). Other areas
with higher productive ability and yields will probably favour the utilitarian role and
timber/non-timber products will be the main focus of forest management. In other remote
inland areas community stability and nature conservation linked with tourist activities is
most likely to be a way to move forward. However, as previously mentioned, all this will
require more effective planning and governance mechanisms able to coordinate different
roles across the area as well as across scales.

7.3. The role of multifunctional forests in sustainable landscapes: wider

implications

This research has examined the ways in which SFM may be guided at different spatial
scales, with emphasis on the cross-scale coordination that is needed in order to implement
landscape planning and governance at the landscape scale. As the landscape scale affords a
territorial framework for sustainable development, it is at this level that multiple
stakeholders should be engaged in order to truly implement multifunctional landscapes. In
a sustainable landscape, multiple functions occur simultaneously and link people and
environment together (Antrop, 2006; Fry, 2001; Jabareen, 2008). Not only in Portugal but
also across Europe, there is a need to look for new, or recreate old, multifunctionalities in
cultural landscapes (Mander et al., 2007), in which different “basins of attraction” might
help to reinforce or create “virtuous circles” through landscapes (Matthews and Selman,
2006).

Forests are only one type of ecosystem and the amount, type and composition of forest
varies hugely across landscapes. As such, forestry is only a part of the problem (Seymour,
2007).

The toolbox used in the context of forestry sector in Portugal enables the:

1. Characterisation of forest conditions at regional scale,

2. Study of the possible roles of forests within a urban-rural gradient,

201
 3. Study of public preferences for forests in contrasting settings,
4. Investigation of possible ways to implement sustainable forestry management by
exploring the type of forestry (ies) of the future.

The work developed in the context of Portugal could be applied elsewhere in Europe and it
is useful to identify the different and possible roles of forestry in contrasting places from
rural to urban regions (see Section 7.3.1). The approach used throughout this work, based
on both quantitative and qualitative approaches, puts emphasis on visual tools for engaging
the public for environmental management (Appleton, 2003). This may be useful in order to
explore the possible roles of forestry in the future. Furthermore, by analysing different
spatial scales, the type of cross-scale coordination required for implementing landscape
scale approaches in Portugal was specified (in this study landscape scale was associated
with the area of river basin catchments).

The set of tools used throughout this work show some ways to move forward, and the
research conducted may be of some use in the contemporary research debate:

1. The approach was innovative in correlating ecological landscape metrics for tree
species with socio-economic indicators for the areas in which forests are located. It
therefore provides an example of a simple interdisciplinary method for
investigating sustainability issues and further demonstrates that landscape metrics
can be used for such a task (Botequilha Leitao and Ahern, 2002).
2. It brought together two of the major issues for sustainable landscapes. Namely, the
provision of goods and services (Potschin and Haines-Young, 2006a; Potschin and
Haines-Young, 2006b) and spatial arrangements (Blaschke, 2006) in order to move
towards SFM.
3. It addressed public preferences for forests in a way that could reinforce the issues in
the ongoing aesthetics-ecology debate (Gobster et al., 2007; Selman, 2006).
4. It further demonstrated the diversity of rural Europe and therefore the need for
diverse approaches to tackle rural development issues (OECD, 2002, 2006). It also
shows that the conditions to effectively tackle the challenge of sustainable forestry
management are not yet in place.

This research shows that there is a need to search for “virtuous circles” (Matthews and
Selman, 2006) which when driven by multiple stakeholder participation are inherently
multifunctional. This multifunctionality reflects a variety of economic activities and social
demands that can replace vicious circles of monofunctionality and dysfunction (Selman,
2006). Though this research may represent a step forward in dealing with SFM across

202
scales, in order to set the frame for the catchment scale, wider and more robust approaches
are needed to fully address the issue of SFM implementation through landscapes. A more
complete approach than the one applied in this study needs to address the issues of
compatibility between the types of forestry (Hersperger, 2006; Hersperger and Forman,
2003), evaluate the goods and services provided by the different types of forestry (Pearce,
2001) at the landscape scale, upscale the results of local scenarios to the watershed level
(de Groot, 2006; MEA, 2003), more effectively present solutions to integrate forests into
formal planning (Selman, 1997), and investigate the types of mechanisms that will enhance
the creation of partnerships amongst different institutions. Figure 7.3 shows other tools
besides those used throughout this study, which may help to study the role of forests in
sustainable landscapes. Above all, the coordination of scales previously referred to in
Section 7.2 must deliver a landscape in which multiple functions occur simultaneously and
link people and environment together. In order to study the different vocations of the
territory, the toolbox used was of value, but it still requires more detailed approaches to
tackle issues that there were not investigated in this research.

However, recognising the need to fully address issues other than those covered by this
research project, it is still possible to draw conclusions regarding the wider implications of
this work. The following sections explore such topics.

203
 EU
Role of multifunctional
forests in sustainable Strengthening of
landscapes environmental policies

Toolbox used Ecosystem National. Portugal

services Integrate forestry into
Landscsape
valuation formal planning
metrics
Scenario
Questionnaire
development Other
survey Visual tools participatory
approaches Regional. Northern
Scenario Portugal
development
Focus Compatibility of Create a “portfolio” of
Multi- group storylines? measures
criteria meetings,
Analysis workshop Land use
and MOLA modelling
Catchment
Conciliate different types
of forestry (ies)

Identify ecological vocations and Municipal/Parish

dynamics in place to implement Define strategies for forest
different vocations management

In a sustainable landscape multiple functions occur

simultaneously and those link people and environment
together

Figure 7.3. The role of multifunctional forests in sustainable landscapes: coordination mechanisms
to deal with it.

7.3.1. European forestry will become more diverse in the future

There is no panacea for forest management! The roles of forests are likely to vary across
environmental conditions as well as socio-economic dynamics. Within the same catchment

204
this study has found that different communities would like to see different “futures” for
forestry in their area of residence. Hence, it was demostrated that different “basins of
attraction” which may deliver contrasting “virtuous circles” (Matthews and Selman, 2006)
are likely to occur even in places with similar characteristics.

This may mean that European forestry will become more diverse in the future, delivering a
mosaic of forest types existing next to each other rather than the homogeneous forestry
pattern of the past (Elands and Wiersum, 2001; Slee, 2007a). It is also likely that across
Europe forestry institutions will put emphasis on good forest management practices such as
continuous cover forestry, managing a mixture of trees and stand ages in order to fulfil
societal demands (Pommerening and Murphy, 2004).

There are already studies which indicate that district heating systems using locally
produced woodchips could produce heat at a lower price than individually heated houses
(Okkonen, 2008) and this might help to attract people to rural areas inverting the trends of
depopulation in rural areas (Firmino, 1999). New technologies are likely to appear for
efficiently transforming forest resources (for example forestry biomass) into renewable
sources of energy and this may bring new roles for forestry in the future. More research
needs to be undertaken in order see the extent to which new uses such as forestry biomass
and carbon sequestration are compatible with SFM practices.

Tourism is likely to increase in the future and forests may play an important role in tourist
activity (Slee et al., 2004). In parallel to these new uses traditional forest management for
production of timber will continue to be important even though the importance of the
protective functions from forests has been increasing (Niskanen and Lin, 2001). The
production system for timber needs to be more environmental friendly than ever, otherwise
it will compromise the outcome of beautiful landscapes for tourism activities
(Pommerening and Murphy, 2004). The problem is to control public demand for timber
products because if Europe decreases internal production of timber it will still need to
import it from elsewhere. If this happens Europe is transferring the problem to other parts
of the world compromising, in this way, sustainability at global scale.

7.3.2. The need for a portfolio of measures for SFM in Europe

As has been shown in the case of Portugal, in order to address such diverse forestry types it
is likely that a portfolio of measures need not only to be put into place but more than that,
the portfolio of measures needs to be efficiently coordinated at global, national and
regional levels (Dwyer, 2007). The problem is that in such different European contexts the
portfolio has to be implemented differentially in different places. For example, in where a
205
community stability discourse is occurring the portfolio of measures will have to be
different from those related to nature conservation (Elands and Wiersum, 2001).
Furthermore the geographical compatibility of the roles of forest needs to be addressed
(Blaschke, 2006). In addition, the goods and services provided in each different type of
forestry (Potschin and Haines-Young, 2006b) need to be placed within broader
geographical settings such as the area of river basins.

There are projects such as RUFUS (www. rufus-eu.de) [Accessed online 25 March 2009 ]
which are already dealing with the diversity of areas in the European context, which may
be used a base for further work. In RUFUS a new typology for rural areas is being created
and local case studies across Europe will investigate the possible futures for rural areas.
This type of approach can be used as a basis for understanding the different roles of
forestry as well as creating a set of policies that will help in the delivery of these futures.

7.3.3. Enhanced communication between researchers, the public and decision

makers is important

The toolbox used throughout this work was based on scientific knowledge from different
research fields and was effective in engaging the general public, forest stakeholders,
decision makers such as parish and municipal councils as well as the managers of the
Peneda-Geres National park. It was clear from the research that there is a need for
partnerships between researchers and decision makers from different backgrounds
(conservation and municipal officers) in order to solve the recurrent dilemma of nature
conservation/biodiversity and sustaining living communities (O' Riordan and Stoll-
Kleemann, 2002) in protected areas across Europe. As such, it is important to investigate
the methods that wider approaches for conservation must be based upon (Margules and
Pressey, 2000).

This study also shows that transdisciplinary approaches are needed to deal with the
multifacets of the concept of multifunctionality (de Groot and Hein, 2007; Fry, 2001;
Mander et al., 2007). It is well known that society is demanding more ecosystem goods and
services from forests such as water and soil protection. However, preferences, for example,
for forests which allow a multitude of recreational activities, may influence the delivery of
other goods and services (Chapter 5). Hence, there is a need to engage with the public in
order to address the issue that some human preferences may undermine the ecology of
forest ecosystems (Gobster et al., 2007; Selman, 2006).

206
7.4 Research caveats and further research

7.4.1. Research caveats: problems and difficulties

This project has been an ambitious attempt at a multidisciplinary project by one person.
Inevitably the results reflect the absence of a research team of experts and some questions
are left unanswered. As was previously discussed in Section 7.2.1., the multi scale
approach was found to be valuable for investigating the roles of forests in sustainable
landscapes, though at different particular scales there are still topics to be explored in order
to fully understand the ways in which forests might contribute to sustainable development
across different regions.

One example concerns the use of landscape metrics. The measures used in this study
(Chapter 3) were informative to distinguishing between industrial and non industrial types
of forestry, but they were not so effective in distinguishing between the pre and post
industrial type of forestry. Furthermore, there are patch metrics, class metrics and
landscape metrics that might be confusing in use. The research used class metrics because
the analysis was conducted at municipality level and these metrics were appropriate for
that scale of analysis. It is known that across Europe the area of municipalities varies from
country to country, necessitating the selection of the appropriate type of metric for each
particular area under analysis. Another problem worth considering is pixel size. In this
study a pixel resolution of 30 meters was used, it is likely however that if the pixel size
were larger or smaller the results could possibly differ. In this study the land cover map
was from the 90s and the socio-economic data used were from the same time period. It is
known that there is a time lag between the socio-economic changes that will reflect on
characteristics of the forests. This study has shown that even with maps and socio-
economic data from the same time period there were differences in forests across the
urban-rural gradient probably because of changes and trends from the past. It is also
worthwhile to point out that in addition to the rural/urban gradient there is also an
environmental gradient across the area and this might be also reflected in the differences in
the forest metrics.

In the analysis at river basin scale both a questionnaire and a field survey were conducted.
One of the problems already pointed out was the small number of plots surveyed in the
field which did not allow a statistical comparison of the characteristics of the forests across
the area. This small number was justified by the difficulties in both 1) obtaining formal
permission to undertake the field survey and 2) obtaining the appropriate help from the
municipal forestry offices to conduct the survey. When initially contacted the municipal

207
forestry offices gave very positive feedback and it was arranged that the municipality
personnel would help in the field. However, when the field survey was about to be
conducted few of the technicians were willing to go to the field on the dates proposed.
Luckily other PhD students offered to help and so the field survey was not cancelled.
Despite not being suitable for conducting statistical analysis the data gathered in the field
survey do indicate that there is a need for more enforced management if moving towards
SFM is the goal (disappearance of old growth forestry and litter in the plots illustrated the
abandonment situation).

The questionnaire survey in 14 municipalities aimed at surveying both landowners and

general public was a very difficult task to carry out. The help of a field assistant was
crucial in completing the survey of 375 people. The method for conducting the survey was
rehearsed between the two interviewers before the start of the survey and the question
cards were read out exactly the same way by each interviewer. Nevertheless, it might be
that other factors not controlled by the researcher had an influence on the results. For
example, in the study of forest fragmentation and simplification (Chapter 5) an attempt was
made to eliminate other confounding factors (e.g. roads) from the photos by varying only
forest cover and stand structure. Even though this effort was made, the ranking of the
photos might be influenced by the presence of blue sky in the photos showing uneven
stand structures while in the even stands was not possible to see the sky due to the
characteristics of the forests. This happened because the researcher felt that realism of the
photos should not be compromised but it is also acknowledged that this might have
influenced the results. Another problem related to the questionnaire survey was managing
the enormous amount of data collected. One of the initial aims of the photo-questionnaire
was to predict public preferences for forests based on socio-economic data such as age, or
relation with forestry (e.g. multiuser, single user). However, this revealed not to be
possible due to the high variability of the data. When ordinal regression models were
evaluated they had a very low pseudo R square and were not much better than the ones
created by chance. This forced a revision of thinking which was time consuming and it was
slow progress to re-plan the structure of Chapter 5.

If Chapter 5 had a slow start, after initiating the research steps of Chapter 6 it was very
difficult to steer the progress of the focus group meetings. The stakeholders involved were
highly motivated and happily continued in the process of meeting at three month intervals
(January, March, May, and July). The problem instead was how not to disappoint such a
motivated group of people. Because the meetings were planned to be carried out every
three months it compromised for example the development of visualisations for the area
208
based on programmes such as Visual Nature Studio (VNS) and instead photo-montages
were used. The problem encountered with the visualisation software was due to the quality
of the data used, especially the 100 meters DEM, which did not allowed sufficient detail in
the topography of the study area. This raised the need to make a trade-off between whether
to spend time in solving the DEM problem (by incorporating additional data) as well as
improving technical skills required for a realistic visualisation or instead to use photo
montages for the area which were quicker to produce. The second option was adopted
which despite not being so sophisticated neatly solved the problem of showing realistic
visualisations of the two different futures. In order to do so the land use models were
imported to the VNS visualisation software and a view point approximately located on the
view point of the photos to show the land use scenarios developed in the IDRISI
programme. Following this, in Photo-Shop the photos were changed in order to show the
different land uses maps in the photos.

The local organisations (Ardal and the forestry office of Ponte da Barca) were of great
support and their involvement in this work was, overall, considered very positive. As
referred to in Chapter 6, the group of participants in the focus group meetings of Gavieira
and Entre Ambos-os-Rios were of different composition and this might have influenced the
choice of different futures for different places. The dynamics created between the
researcher and the stakeholders initiated an attempt to implement the two storylines via a
pilot project. This was not an initial goal but the researcher felt the necessity of not
disappointing the stakeholders‟ aspirations. For this, an expert meeting was organised and
despite acknowledging the validity of the work, the forestry institutions (the National
Forestry Office-DGRF and Peneda Geres National Park) were not willing to test the
implementation of the two very different strategies for forestry management. Despite this
the National Park did request the information related to the storylines and included these in
the management plan of the National Park.

A problem experienced during the research was the fact that it was based in England
addressing sustainable forestry management in Portugal. Dealing with this situation
required coordination of tasks and the setting of a rigid calendar. With obvious problems to
face such as programming field work and meetings in Portugal while based in England, of
foremost importance was the network of contacts that I had from seven years of experience
as a forestry engineer in a non-governmental institution in Northern Portugal. The contacts
already in place were crucial to the successful development of the different research steps.

209
7.4. 2. Future research: there is no panacea but there might be trends...

The type of functions from forests likely to be “explored” in different SESs is likely to
depend upon socio-economic and environmental characteristics of the regions. Although
acknowledging the no panacea rule (Ostrom, 2007; Ostrom et al., 2007) it will be
challenging, though possible, to investigate which forest functions may work as “basins of
attraction” in different places across Europe. For example, by using some of different
groups of countries created by Slee (2007a), it would be feasible to undertake case studies
based on participatory approaches such as focus group meetings (that may be framed by
similar questions as the ones addressed in Chapter 6) in different countries across Europe.
A case study approach at the local level would be ideal, though time consuming, also
requiring a huge amount of resources, however, depicting the possible roles of forests
could also be inferred from a literature review of participatory studies of natural resource
management across Europe.

The main aim of this future work would be to investigate which forest functions may work
as “basins of attraction” in different places (Matthews and Selman, 2006). The aggregation
of the results of the local focus groups (or from reviewing research done so far) would give
a picture of the type of forestry in the future and the ways in which it is likely that they will
be distributed across Europe. These forestry types have to be translated into land cover
maps. Local land use cover patterns then have to be upscaled for example to the area of
watershed catchments in order to study the multiple forest functions at the landscape level
(de Groot, 2006). One approach that could be used is to calculate landscape metrics.

As previously discussed in Chapter 3 the set of metrics used throughout this work was not
so effective in distinguishing the pre and post industrial types of forestry so, more metrics
would need to be included. It would be useful to have a set of landscape metrics able to
distinguishing different types of forestry. Another approach would be to explore how the
metrics for tree species in an area are correlated with the surrounding land uses. For
example, by using available land use maps such as CORINE 2000 to calculate a set of
metrics characterising, agricultural, urban and forests to see what type of correlations there
are. Following this, the functions associated with them could be investigated. For example,
whether a forest with particular composition when surrounded by x,y,z has associated
a,b,c functions with it. This will allow a better understanding of the role of forests in
different places.

Such an approach would allow the study of possible roles of forests across regions (de
Groot, 2006). This will certainly help in framing the approaches that will be required for

210
planning of sustainable landscapes in Europe. In some areas, for example in the
Mediterranean, likely issues that policies need to tackle relate to water shortages that are
likely to be important in the future. As such, the type of forestry favouring this service (e.g.
continuous cover forestry) will have to develop mechanisms (e.g. for effective carbon
credit markets) that will incentivise forest landowners in going for such as strategy.
Already in the present and certainly in the future it is likely that policies subsidizing
forestry and agriculture will not be viable, instead investments (OECD, 2006) need to be
made to re-create virtuous circles (Matthews and Selman, 2006).

7.5 Concluding remarks

Forests inhabited the earth well before humans did. SD and SFM imply that humans will
be able to manage the forests in such a manner that they will benefit from the multitude of
functions that forests are able to provide. We will arrive there....hopefully!

Implementing SD through landscapes has been proposed as a basis for moving forward and
this has been addressed in research agendas that are increasingly engaging with a broad
group of stakeholders through transdisciplinarity approaches. Fulfilling the demands of a
broad group of stakeholders requires managing to obtain more than one function from a
plot of land thus highlighting the need to create multifunctional landscapes. Forests
provide a multitude of functions that are essential to achieving this but the forest
contribution is only a part of the landscape sustainability.

As a concluding remark it is fair to say that the work developed throughout this PhD has
shown some ways to move forward towards sustainable forestry management that might
contribute to more sustainable landscapes in the context of Portugal. The toolbox used,
based on public consultation aided by visual tools and GIS techniques, has proven to be
robust for addressing sustainable forestry management and in exploring the ways in which
forestry is in accordance with the dynamics of the whole landscape. This multi scale
approach was able to explore the ways in which cross-scale coordination can be made in
order to put into practice a landscape scale approach. As this research shows, there is still a
need for improving and coordinating planning and governance systems in order to pursue
sustainability across a range of scales. There is also a need to reconcile top down and
bottom up approaches to reach the overarching goal of pursing sustainable development
through landscapes. Finally, there is a need to continuously look for measures that may
create virtuous circles in which economy, society and environment are well balanced. The
challenges ahead are enormous thus the objectives may seem utopian, but there is evidence
that pursuing SD and SFM through a landscape perspective is actually attainable.

211
References
Abello, R. P., Bernaldez, 1986, Landscape preference and personality, Landscape and
Urban Planning 13:19-28.

Adger, W. N., Brown, K., Jorban, A., Paavola, J., Rosendo, S., Seyfang, G., 2004,
Governance for sustainabilty: Towards a thick undesrtanding of evironmental decision
making, CSERGE Working paper EDM 02-04
(http://www.uea.ac.uk/env/cserge/pub/wp/by_year.htm (March 2009).

Ahern, J., 1995, Greenways as a planning strategy, Landscape and Urban Planning 33(1-
3):131-155.

Ahern, J., 2000, Sustainable urban landscapes: The Surrey design charrette, Landscape and
Urban Planning 50(4):271-273.

Ahern, J., 2005, Theories, Methods & Strategies for Sustainable Landscape Planning, in:
From landscape research to landscape planning: Aspects of integration, education and
application (B. Tress, Tress, G., Fry, G., Opdam, P. , ed.), Springer, Wagenigen, pp. 119-
131.

Akselsson, C., Westling, O., Sverdrup, H., Gundersen, P., 2007, Nutrient and carbon
budgets in forest soils as decision support in sustainable forest management, Forest
Ecology and Management 238 167–174.

Andresen, M. T., Castelbranco, C., 1993, Heading for a Postmodern Landscape -

Portuguese Trends in the Reconciliation of Environmental-Quality and Landscape
Planning with Economic-Development, Landscape and Urban Planning 23(3-4):183-194.

Angelstam, P., Kapylova, E., Korn, H., Lazdinis, M., Sayer, J. A., Teplyakov, V.,
Tornblom, J., 2005, Changing Forest values in Europe, in: Forest in Landscapes.
Ecosystem Approaches to Sustainability (J. A. Sayer, S. Maginnis, eds.), Earthscan,
London, pp. 59-74.

Antrop, M., 2004, Landscape change and the urbanization process in Europe, Landscape
and Urban Planning 67(1-4):9-26.

Antrop, M., 2005, Why landscapes of the past are important for the future?, Landscape and
Urban Planning 70(1-2):21-34.

212
Antrop, M., 2006, Sustainable landscapes: contradiction, fiction or utopia?, Landscape and
Urban Planning 75(3-4):187-197.

Antunes, P., Santos, R., Videira, N., 2006, Participatory decision making for sustainable
development-the use of mediated modelling techniques, Land Use Policy 23:44-52.

Appleton, K., Lovett, A., 2003, GIS-based visualisation of rural landscapes: defining
'sufficient' realism for environmental decision-making, Landscape and Urban Planning
65(3):117-131.

Appleton, K. J., 2003, GIS-Based landscape visualisation for environmental management,

PhD Thesis, in: School of Environmental Sciences, University of East anglia, Norwich, pp.
168.

Araújo, J. F. F. E., 2001, Developments in Change Management in Portuguese Central

Administration: the dilemma between hierarchy and agencification (E. d. E. e. G.
Universidade do Minho, Campus Universitário de Gualtar, P. BRAGA, eds.), Universidade
do Minho, Escola de Economia e Gestão, Campus Universitário de Gualtar4710-057
BRAGA, PORTUGAL, BRAGA, PORTUGAL.

Aspinal, R., Pearson, D., 2000, Integrated geographical assessement of environmental

condition in water catchments: linking landscape ecology, environmental modelling and
GIS, Journal of Environmental Management 59:299-319.

Balmford, A., Bruner, A., Cooper, P., Costanza, R., Farber, S., Green, R. E., Jenkins, M.,
Jefferiss, P., Jessamy, V., Madden, J., Munro, K., Myers, N., Naeem, S., Paavola, J.,
Rayment, M., Rosendo, S., Roughgarden, J., Trumper, K., Turner, R. K., 2002, Ecology -
Economic reasons for conserving wild nature, Science 297(5583):950-953.

Baskent, E. Z., 2007, Reflection on the current status and future challenges of forest
management planning system in Turkey, Scientific Tools and Research Needs for
Multifunctional Mediterranean Forest Ecosystem Management (56):35-43

Baskent, E. Z., Jordan, G. A., Nurullah, A. M. M., 2000, Designing forest landscape
management, Forestry Chronicle 76(5):739-742.

Baskent, E. Z., Yolasigmaz, H. A., 1999, Forest landscape management revisited,

Environmental Management 24(4):437-448.

Baskent, E. Z., Yolasigmaz, H. A., 2000, Exploring the concept of a forest landscape
management paradigm, Turk J Agric For 24:443–451.
213
Basu, P., 2009, A green investment, Nature 457(8 January).

Bateman, I. J., Brouwer, R., Davies, H., Day, B. H., Deflandre, A., Di Falco, S., Georgiou,
S., Hadley, D., Hutchins, M., Jones, A. P., Kay, D., Leeks, G., Lewis, M., Lovett, A. A.,
Neal, C., Posen, P., Rigby, D., Turner, R. K., 2006, Analysing the agricultural costs and
non-market benefits of implementing the Water Framework Directive, Journal of
Agricultural Economics 57(2):221-237.

Bateman, I. J., Lovett, A., 2000, Modelling and valuing carbon sequestration in softwood
and hardwood tres, timber products and forest soils CSERGE Working paper, University of
East Anglia GEC 2000-13 (http://www.uea.ac.uk/env/cserge/pub/wp/by_year.htm (March
2009)).

Bateman, I. J., Lovett, A. A., Brainard, J. S., 2005, Applied Environmental economics: a
GIS approach to Cost-benefit Analysis, Cambridge University Press, Cambridge, UK.

Blaschke, T., 2006, The role of the spatial dimension within the framework of sustainable
landscapes and natural capital, Landscape and Urban Planning 75(3-4):198-226.

Bogaert, J., 2002, Forests and landscapes-linking ecology, sustainability and asthetics
S.R.J.Sheppard, H.W. Harshaw (Eds), IUFRO research series 6, CABI Publishing in
association with IUFRO 2001, Landscape and Urban Planning 59:125-127.

Bolund, P., Hunhammar, S., 1999, Ecosystem services in urban areas, Ecological
econmomics 29:293-301.

Borjeson, L., Hojer, M., Dreborg, K. H., Ekval, T., Finnveden, G., 2006, Scenario types
and techniques: Towards a user's guide, Futures 38:723-739.

Botequilha Leitao, A., Ahern, J., 2002, Applying landscape ecological concepts and
metrics in sustainable landscape planning, Landscape and Urban Planning 59(2):65-93.

Boyle, G., 2004, Renewable energy- Power for a sustainable future, in: Renewable energy-
Power for a sustainable future (G. Boyle, ed.), Oxford University Press in association with
Open University, Oxford.

Brainard, J., Bateman, I., Lovett, A., 2001, Modelling demand for recreation in English
woodlands, Forestry 74(5):423-438.

Brandt, J., Tress, B., Tress, G., 2000, Multifunctional landscapes:Interdisciplinary

approaches to landscape research and management (J. Brandt, B. Tress, G. Tress, eds.),

214
Centre for landscape research Roscilde, Denmark, Roskilde, Denmark, October 18-20-
2000.

Brouwer, R., 1999, Changing name-tags: A legal antropolological approach to communal

lands in Portugal, Journal of Legal Pluralism 43,:1-30.

Canas, A., Ferrao, P., Conceicao, P., 2003, A new environmental Kuznets curve?
Relationship between direct material input and income per capita: evidence from
industrialised countries, Ecological Economics 46(2):217-229.

Carlman, I., 2005, The rule of sustainability and planning adaptativity, Ambio 34(2):163-
168.

Carpenter, S., Benett, E. M., Peterson, G., 2006, Scenarios for ecosystem services: An
overview, Ecology & Society 11(1):[online] URL:
http://ecologyandsociety.org/vol11/iss1/art29/.

Carvalho-Ribeiro, S. M., Lovett, A., 2009, Associations between forest characteristics and
socio-economic development: A case study from Portugal, Journal of Environmental
Management (in press).

Cash, D. W., Adger, W. N., Berkes, F., Garden, P., Lebel, L., Olsson, P., Pritchard, L.,
Young, O., 2006, Scale and cross-scale dynamics: Governance and Information in a
multilevel World, Ecology & Society 11(2):[online] URL
http://ecologyandsociety.org/vol11/iss2/art8/.

Cash, D. W., clark, W. C., Alcocock, F., Dickson, N. M., Eckley, N., Guston, D. H., Jager,
J., Mitchell, R. B., 2003, Knowledge systems for sustainable development, Proceedings of
the National Academy of Sciences of the United States of America 100(14):8086-8091.

Cashman, A., 2006, A watery form of sustainability, Water and Environment journal 20:2-
6.

Catry, F., Damasceno, P., Silva, J. S., Galante, M., Moreira, F., 2005, Spatial distribution
patterns of wildfire ignitions in Portugal, in: Wildfire 2007, Sevilla, Spain.

CBD, 2008, Cross-sectoral toolkit for the conservation and sustainable management of
forest biodiversity, in: Technical Series No.39 (I. Thompson, T. Christophersen, eds.),
UNEP, FAO, CBD, Montreal, Canada.

215
CEMAT, 2007, Spatial development glossary, in: Territory and landscape n. 2, Council of
Europe Publishing, Strasbourg.

Clemente, A., Rego, F., Correia, O., 2005, Growth water relations and photosynthesis of
seedlings and resprouts after fire, Acta Oecologica 27:233-243.

Costanza, R., Daly, H. E., 1992, Natural capital and sustainable development,
Conservation Biology 6(1):37-46.

Costanza, R., dArge, R., deGroot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K.,
Naeem, S., ONeill, R. V., Paruelo, J., Raskin, R. G., Sutton, P., vandenBelt, M., 1997, The
value of the world's ecosystem services and natural capital, Nature 387(6630):253-260.

Council of Europe, 2000, The European Landscape Convention, Strasbourg.

Daniel, T. C., 2001, Whither scenic beauty? Visual landscape quality assessement in the 21
st century, Landscape and Urban Planning 54:267-281.

Daniel, T. C., Meitner, M. M., 2001, Representational validity of landscape visualisations:

The effects of graphical realism on perceived scenic beauty of forest vistas, Journal of
environmental Psychology 21(61-72).

de Blust, G., Olmen, M. V., 2000, Monitoring multifunctional terrestrial landscapes: some
coments, in: Multifunctional landscapes: Interdisciplinary Approaches to Landscape
Reseach and Management (J. Brandt, B. Tress, G. Tress, eds.), Centre for Landscape
Research, Roskilde, Denmark, Centre for Landscape Research, Roskilde, Denmark.

de Groot, R., 2006, Function-analysis and valuation as a tool to assess land use conflicts in
planning for sustainable, multifunctional landscapes, Landscape and Urban Planning
75:175-186.

de Groot, R., Hein, L., 2007, Concepts and valuation of landscape functions at different
scales, in: Multifunctional land use: meeting future demands for landscape goods and
services (U. Mander, H. Wiggering, K. Helming, eds.), Springer, Berlin, pp. 14-36.

de Groot, R. S., Wilson, M. A., Boumans, R. M. J., 2002, A typology for the classification,
description and valuation of ecosystem functions, goods and services, Ecological
Economics. 41(SPECIAL ISSUE: The Dynamics and Value of Ecosystem Services:
Integrating Economic and Ecological Perspectives):393–408.

216
DEFRA, 2006, Valuing our natural environment, in: Final report NR0103, Department for
Environment, Food and Rural Affairs, London.

Deybe, D., 2007, Policies, research perspective and challenges on multifunctional land use,
in: Multifunctional land use. Meeting future demands for landscape goods and services (U.
Mander, H. Wiggering, K. Helming, eds.), Springer, Berlin, pp. 143-152.

DGF, 1996, Lei de bases da política florestal., in: Lei n. 33/96 (D. g. d. florestas, ed.),
Direcção-Geral das Florestas, Lisbon, pp. 3-5.

DGRF, 2007a, Inventario Florestal Nacional, in: http://www.dgrf.min-agricultura.pt/portal

(September 2008), Direcao Geral dos Recursos Florestais, Lisbon.

DGRF, 2007b, Matriz estruturante para as florestas, in: http://www.dgrf.min-

agricultura.pt/portal (September 2008) (DGRF, ed.), Direccao Geral dos Recursos
Florestais, Lisbon.

DiBari, J. N., 2007, Evaluation of five landscape-level metrics for measuring the effects of
urbanisation on landscape structure: the case of Tucson, Arisona, USA, Landscape and
Urban Planning 79(3-4):308-313.

Dolman, P. M., Lovett, A., O'Riordan, T., Cobb, D., 2001, Designing Whole Landscapes,
Landscape Research 26(4):305-335.

DTI, 2007, Economic Analysis of biomass energy, Department of Trade and Industry, in:
UK Biomass Strategy (DTI, ed.), Energy Technologies Unit: Department of Trade and
Industry, London.

Duhme, F., Pauleit, S., Baier, H., 1997, Quantyfying targets for nature conservation in
future European landscapes, Landscape and Urban Planning 37:73-84.

Dwyer, J., 2007, The state of the rural environment in Europe:what challenges and
oportunities for future policies, in: Future policies for rural Europe 2013 and beyond-
delivering sustainable rural land management in a changing Europe (L. u. P. Group, ed.),
Brussels.

Eastman, J. R., 2006, IDRISI Andes Manual, Clark Labs, Clark University. Worcester.
MA. USA.

Ehrlich, P., 1996, Conservation in temperate forests:what do we need to know and do?,
Forestry Ecology and Management 85:9-19.

217
Elands, B. H. M., O'Leary, T. N., Boerwinkel, H. W. J., Wiersum, K. F., 2004, Forest as a
mirror of rural conditions; local views on the role of forests across Europe, Forest Policy
and Economics 6:469-482.

Elands, B. H. M., Praestholm, S., 2008, Landowners' perspectives on the rural future and
the role of forests across Europe, Journal of Rural Studies 24(1):72-85.

Elands, B. H. M., Wiersum, K. F., 2001, Forestry and rural development in Europe: an
exploration of socio-political discourses, Forest Policy and Economics 3(1-2):5-16.

ELC, 2000, The European Landscape Convention (C. o. Europe, ed.), Strasbourg.

Evernden, N., 1988, The Beauty of the Environment - a General-Model for Environmental
Aesthetics - Sepanmaa,Y, Environmental Ethics 10(2):183-186.

Fahrig, L., 2003, Effects of habitat fragmentation on biodiversity, Annual Review Ecology
and Evolution 34:483-515.

FAO, 2003, Sustainable forest management and the ecosystems approach: two concepts
one goal, in: Forest management working paper FM 25 (F. f. department, ed.), FAO,
Rome.

FAO, 2005, Global forest resources assessement update, in: Working paper 83 (FAO, ed.),
Food and Agriculture Organization of the United Nations, Rome.

Fernandes, P. A. M., 2001, Fire spread prediction in shrub fuels in Portugal, Forest
Ecology and Management 144:67-74.

Fidelis, T., Sumares, D., 2008, Nature conservation and urban development control in the
Portuguese planning system: A new impetus agains old praxis?, European Environment
18:298-311.

Firmino, A., 1999, Agriculture and landscape in Portugal, Landscape and Urban Planning
46(1-3):83-91.

Fisher, B., Costanza, R., Turner, K., Morling, P., 2004, Defining and Classifying
Ecosystem Services for Decision Making, in: CSERGE Working Paper EDM 07-04
(CSERGE, ed.), UEA, Norwich.

Flick, U., 2002, An Introduction to qualitative research (Sage, ed.), London.

218
Foley, J. A., Asner, G. P., Costa, M. H., Coe, M. T., DeFries, R., Gibbs, H. K., Howard, E.
A., Olson, S., Patz, J., Ramankutty, N., Snyder, P., 2007, Amazonia revealed: forest
degradation and loss of ecosystem goods and services in the Amazon Basin, Frontiers in
Ecology and the Environment 5(1):25-32.

Folke, C., Carpenter, S., Elmqvist, T., Gunderson, L., Holling, C. S., Walker, B., 2002,
Resilience and Sustainable Development: Building Adaptive Capacity in a World of
Transformations, Ambio 31(5):437-440.

Folke, C., Hahan, T., Olsson, P., Norberg, J., 2005, Adaptative governance of social-
ecological systems, Annual Review Environmental Resources 30:8.1-8.33.

Fonseca, P. A. L., 2000, Indices de desenvolvimento concelhio, in: Volume II. 2.

quadrimestre de 2002. (I. N. d. Estatistica, ed.), Instituto Nacional de Estatistica, Lisbon,
pp. 1-34.

Forman, R. T. T., 1995, Land mosaics: the ecology of landscapes and regions, Cambridge
University Press, Cambridge, pp. 632.

Forman, R. T. T., Collinge, S. K., 1997, Nature conserved in changing landscapes with and
without spatial planning, Landscape and Urban Planning 37(1-2):129-135.

Fry, G., 2001, Multifunctional landscapes-towards transdiciplinary research, Landscape

and Urban Planning 159:159-168.

Gamborg, C., Larsen, J., 2003, Back to nature - a sustainable future for forestry?, Forest
Ecology and Management 179,:559-571.

Gibbons, P., Lindenmayer, D. B., Fischer, J., Manning, A. D., Weinberg, A., Seddon, J.,
Ryan, P., Barrett, G., 2008, The Future of Scattered Trees in Agricultural Landscapes,
Conservation Biology 22(5):1309-1319.

Gobster, P. H., Nassauer, J. I., Daniel, T. C., Fry, G., 2007, The shared landscape: what
does aesthetics have to do with ecology?, Landscape Ecology 22:959-972.

Green, R. E., Cornell, S. J., Scharlemann, J. P. W., Balmford, A., 2005, Farming and the
fate of wild nature, Science 307(28 January):550-555.

Grimble, R., Chan, M. K., 1995, Stakeholder analysis for natural resources management in
developing countries, Natural Resource Forum 19(2):113-124.

219
Hagedorn, K., 2007, Towards an institutional theory of multifunctionality, in:
Multifunctional land use. Meeting future demands for landscape goods and services (U.
Mander, H. Wiggering, K. Helming, eds.), Springer, Berlin, pp. 105-124.

Hanna, K. S., 2005, Planning for sustainability, Journal of the American Planning
Association 71(1):27-40.

Harshaw, H. W., Kozak, R. A., Sheppard, S. R. J., 2006, How well are outdoor
recreationists represented in forest land use planning? Perceptions of recreationists in the
Sea-toSky corridor of British Columbia, Landscape and Urban Planning 78:33-49.

Harshaw, H. W., Tindall, D. B., 2005, Social structure, identities and values: A network
approach to undestanding people's relationships to forests, Journal of Leisure Research
37(4):426-449.

Hawkins, V., Selman, P., 2002, Landscape scale planning: exploring alternative land use
scenarios, Landscape and Urban Planning 60(4):211-224.

Hector, A., Bagchi, R., 2007, Biodiversityand ecosystem multifunctionality, Nature 448(12
July).

Hedblom, M., Soderstrom, B., 2008, Woodlands across Swedish urban gradients: Status,
structure and management implications, Landscape and Urban Planning 84,:62-73.

Herrmann, S., Osinski, E., 1999, Planning sustainable land use in rural areas at different
spatial levels using GIS and modelling tools, Landscape and Urban Planning 46:93-101.

Hersperger, A. M., 2006, Spatial adjencies and interactions: Neighborhood mosaics for
landscape ecologycal planning, Landscape and Urban Planning 77:227-239.

Hersperger, A. M., Forman, R. T. T., 2003, Adjacency arrangement effects on plant

diversity and composition in woodland patches, Oikos 101(2):279-290.

Higman, S., Bass, S., Judd, N., Mayers, J., Nussbaum, R., 1999, The sustainable forestry
handbook- A pratical guide for tropical forest managers on implementing new standards
(E. p. Ltd, ed.), London.

IGEO, 1990, Carta de ocupação do solo (COS' 90) (I. G. d. Portugal, ed.), Instituto
Geografico de Portugal, Lisbon.

INAG, 2005, Planos de Bacia Hidrografica, in: Instituto da Agua, INAG, Portugal,
http://www.inag.pt/ (March 2009), Lisbon.
220
INAG, 2006, Plano Nacional da Agua, in: Instituto da Agua, INAG, Portugal,
http://www.inag.pt/ (March 2009), Lisbon.

INE, 2001, Retractos territoriais (I. N. d. E. (Portugal), ed.), Instituto Nacional de

Estatística (Portugal), Lisbon.

INE, 2008, Estatisticas Demograficas 2007, in: Populacao e Sociedade (I. N. d. E. INE,
ed.), Instituo Nacional de Estatistica INE, LISBON.

Jabareen, Y. R., 2008, A new conceptual framework for sustainable development,

Environment Development and Sustainability (10):179-192.

Jacobs, P., 1986, Sustaining landscapes: sustaining societies, Landscape and Urban
Planning 13:349-358.

Janse, G., Ottitsch, A., 2005, Factors influencing the role of Non-Wood Forest Products
and Services, Forest Policy and Economics 7(3):309-319.

Ji, W., Ma, J., Twibell, R., Underhill, K., 2006, Characterising urban sprawl using multi-
stage remote sensing images and landscape metrics, Computers, Environment and Urban
Systems 30:861-879.

Johnson, N., White, A., Perrot-Maite, D., 2002, Developing markets for water services
from forests: Issues and lessons for innovators (F. trends, W. R. Institute, T. K. group,
eds.), USA.

Jongman, R. H. G., 2002, Homogenisation and fragmentation of the European landscape:

ecological consequences and solutions, Landscape and Urban Planning 58(2-4):211-221.

Kangas, J., Kangas, A., 2005, Multiple criteria decision support in forest management- the
approach, methods applied and experiences gained, Forest Ecology and Management
207(1-2):133-143.

Karjalainen, E., Komulainen, M., 1998, Field afforestation preferences: A case study in
northeastern Finland, Landscape and Urban Planning 43:79-90.

Karjalainen, T., Pussinen, A., Liski, J., Nabuurs, G. J., Eggers, T., Lapvetelainen, T.,
Kaipainen, T., 2003, Scenario analysis of the impacts of forest management and climate
change on the European forest sector carbon budget, Forest Policy and Economics 5:141-
155.

221
Kato, S., Ahern, J., 2008, Learning by doing: adaptative planning as a strategy to adress
uncertainty in planning, Journal of Environmental Planning and Management 51(4):543-
559.

Kellomaki, S., Pukkala, T., 1989, Forest landscape: A method of amenity evaluation based
on computer simulation, Landscape and Urban Planning 18:117-125.

Knickel, K., Renting, H., 2000, Methodological and conceptual Issues in the study of
multifunctionality and rural development, Sociologia Ruralis 40(4):512-528.

Kok, K., Rothmanc, D. S., Patel, M., 2006, Multi-scale narratives from an IA perspective:
Part I.

European and Mediterranean scenario development, Futures 38:261–284.

Kozlowski, J. I., Hill, G., 1993, Towards planning for sustainable development. A guide
for the ultimate environmental treshold (UET) method (J. I. Kozlowski, G. Hill, eds.),
Avebury, Brookfield USA, pp. 373.

Kummar, S., Kant, S., 2007, Exploded logit modelling of stakeholders' preferences for
multiple forest values, Forest Policy and Economics 9:516-526.

Lee, N. K., 1995, Deliberately seeking sustainability in the Columbia river basin, in:
Barriers and bridges to the renewal of ecosystems and institutions (C. S. Holling, S. S.
Light, eds.), Columbia University Press, New-York, pp. 214-238.

Lele, S. M., 1991, Sustainable development- a critical review, World Development

19(6):607-621.

Leskinen, L. A., Leskinen, P., Kurttila, M., Kangas, J., Kajanus, M., 2006, Adapting
modem strategic decision support tools in the participatory strategy process - a case study
of a forest research station, Forest Policy and Economics 8(3):267-278.

Lewis, J. L., Sheppard, S. R. J., 2006, Culture and comunication: Can landscape
visualisation improve forest management consultation with indigenous comunities?,
Landscape and Urban Planning 77:291-313.

Lexer, M. J., Brooks, R. T., 2005, Decision support for multple purpose forestry, Forest
Ecology and Management 207:1-3.

222
Lexer, M. J., Honninger, K., Scheifinger, H., Matulla, C., Groll, N., Kromp-Kolb, H.,
2000, The sensitivity of central European mountain forests to scenarios of climatic change:
Methodological frame for a large-scale risk assessment, Silva Fennica 34(2):113-129.

Lindenmayer, D., Hobbs, R. J., Montague-Drake, R., Alexandra, J., Bennett, A., Burgman,
M., Cale, P., Calhoun, A., Cramer, V., Cullen, P., Driscoll, D., Fahrig, L., Fischer, J.,
Franklin, J., Haila, Y., Hunter, M., Gibbons, P., Lake, S., Luck, G., MacGregor, C.,
McIntyre, S., Nally, R. M., Manning, A., Miller, J., Mooney, H., Noss, R., Possingham, H.,
Saunders, D., Schmiegelow, F., Scott, M., Simberloff, D., Sisk, T., Tabor, G., Walker, B.,
Wiens, J., Woinarski, J., Zavaleta, E., 2008, A checklist for ecological management of
landscapes for conservation, Ecology Letters 11:78-91.

Lindenmayer, D. B., 1999, Future directions for biodiversity conservation in managed

forests:indicator species, impact studies and monitoring programs, Forest Ecology and
Management (115):277-287.

Liu, J., Taylor, W., 2002, Coupling landscape ecology with natural resource management:
Paradigm shifts and new approaches, in: Integrating landscae ecology into natural
resource management (J. Liu, W. Taylor, eds.), Cambridge University Press, Cambridge,
UK.

Lopes-Ridaura, S., Keulen, H. V., Ittersum, M. V., Leffelaar, P. A., 2005, Multiscale
methodologycal Framework to derive criteria and indicators for sustainability evaluation of
peasant natural resource management systems, Environment Development and
Sustainability 7:51-69.

Malczewski, J., 1999, GIS and Multicriteria Decision Analysis. (I. John Wiley & Sons,
ed.), USA.

Malczewski, J., 2004, GIS-based land-use suitability analysis: a critical overview,

Progress in Planning 62:3-65.

Mander, U., Helming, K., Wiggering, H., 2007, Multifunctional land use: meeting future
demands for landscape goods and services, in: Multifunctional land use: meeting future
demands for landscape goods and services (U. Mander, H. Wiggering, K. Helming, (Eds),
eds.), Springer, Berlin.

Mangold, R. D., 1995, Sustainable development- the forest services approach, Journal of
forestry 93(11):25-28.

223
MAOTDR, 2007, Speech delivered by the Portuguese Minister of Environment, Territorial
Planning and Regional Development Francisco Nunes Correia, in: European Water
Conference (E. C. s. E. Directorate-General, ed.), Brussels.

Margules, C. R., Pressey, R. L., 2000, Systematic conservation planning, Nature 405(11
May):243-253.

Mather, A. S., 1992, The forest transition, Area 24(4):367-379.

Mather, A. S., 2004, Forest transition theory and the reforesting of Scotland, Scottish
Geographical Journal 120(1-2):83-98.

Mather, A. S., Fairbairn, J., Needle, C. L., 1999, The course and drivers of the forest
transition: the case of France, Journal of Rural Studies 15(1):65-90.

Mather, A. S., Needle, C. L., 1998, The forest transition: a theoretical basis, Area
30(2):117-124.

Mather, A. S., Needle, C. L., Coull, J. R., 1998, From resource crisis to sustainability: the
forest transition in Denmark, International Journal of Sustainable Development and World
Ecology 5(3):182-193.

Matsuoka, R. H., Kaplan, R., 2008, People needs in the urban landscape: Analysis of
Landscape and Urban Planning contributions, Landscape and Urban Planning 84:7-19.

Matthews, R., Selman, P., 2006, Landscape as a focus for integrating human and
environmental processes, Journal of Agricultural Economics 57(2):199-212.

McGarigal, K., Cushman, S. A., Neel, M. C., Ene, E., 2002, FRAGSTATS: Spatial pattern
analysis program for categorical maps, Computer software program produced by the
authors at the University of Massachusetts, Amherst.,
www.umass.edu/landeco/research/fragstats/fragstats.htm, 2009/03/09.

McIntyre, S., Hobbs, R., 1999, A framework for conceptualizing human effects on
landscapes and its relevance to management and research models, Conservation Biology
13(6):1282-1292.

McMichael, A. J., Butler, C. D., Folke, C., 2003, New visions for adressing sustainability,
Science 302(12 December):1919-1920.

MCPFE, 2007, Ministerial Conference for protection of forests in Europe, in:

http://www.mcpfe.org/.
224
MEA, 2003, The multiscale approach. Chapter 4, in: Millenium Ecosystems Assessement,
http://www.millenniumassessment.org/en/index.aspx (9 March 2009), pp. 61-83.

MEA, 2005, Synthesis Reports: Biodiversity, in:

http://www.millenniumassessment.org/en/index.aspx ( 10 January 2009).

Meitner, M. J., Sheppard, S. R. J., Cavens, D., Gandy, R., Picard, P., Harshaw, H.,
Harrison, D., 2005, The multiple roles of environmental data visialisation in evaluating
alternative forest management strategies, Computers and Electronics in Agriculture
49:192-205.

Mendoza, G., Prabhu, R., 2005, Combining participatory modelling and multicriteria
analysis for comunity-based forest management, Forest Ecology and Management
207:145-156.

Milligan, J., O'Riordan, T., 2007, Governance for sustainable coastal futures, Coastal
Management 35:499-509.

Milligan, J., O'Riordan, T., Nicholson-Cole, S. A., Watkinson, A. R., 2009, Nature
conservation for future sustainable shorelines: lessons from seeking to involve the public,
Land Use Policy 26:206-213.

Misgav, A., 2000, Visual preference of the public for vegetation groups in Israel,
Landscape and Urban Planning 48(3-4):143-159.

Moore, S., Wallington, T., Hobbs, R., Ehrlich, P., Holling, C., Levin, S., Lindenmayer, D.,
Pahl-Wostl, C., Possingham, H., Turner, M., Westoby, M., 2009, Diversity in Current
Ecological Thinking: Implications for Environmental Management, Environmental
Management 43(1):17-27.

Moreira, F., Ferreira, P. G., Rego, F. C., Bunting, S., 2001a, Landscape changes and
breeding bird assemblages in northwestern Portugal: the role of fire, Landscape Ecology
16(2):175-187.

Moreira, F., Rego, F. C., Ferreira, P. G., 2001b, Temporal (1958-1995) pattern of change
in a cultural landscape of northwestern Portugal: implications for fire occurrence,
Landscape Ecology 16(6):557-567.

Munda, G., 2005, Measuring sustainability: a multi-criterion framework, Environment

Development and Sustainability 7:117-134.

225
Nabuurs, G. J., Paivinen, R., Schanz, H., 2001, Sustainable management regimes for
Europe's forests- a projection with EFISCEN until 2050, Forest Policy and Economics
3:155-173.

Nagendra, H., 2007, Drivers of reforestation in human dominated forests, Proceedings of

the National Academy of Sciences of the United States of America 104(39):15218-15223.

Naveh, Z., 2007, Landscape ecology and sustainability, Landscape Ecology 22(10):1437-
1440.

Nicholson-Cole, S. A., O'Riordan, T., 2009, Adaptive governance for a changing coastline:
science, policy and publics in search of a sustainable future, in: Adapting to Climate
Change: Thresholds, Values, Governance (W. N. Adger, I. Lorenzoni, K. O'Brien, eds.),
Cambridge University Press, Cambridge, UK.

Niskanen, A., Lin, C., 2001, Regional similarities of forest resources and socio-economic
structures in EU member states, Forest Policy and Economics 3:55-67.

Niza, S., Ferrao, P., 2006, A transitional economy's metabolism: The case of Portugal,
Resources Conservation and Recycling 46(3):265-280.

Noss, R. F., 1999, Assessing and monitoring forest biodiversity: A suggested framework
and indicators, Forest Ecology and Management 115(2-3):135-146.

O' Riordan, T., Stoll-Kleemann, 2002, Biodiversity, sustainability and human comunities.
Protecting beyond the protected (T. O'Riordan, Stoll-Kleemann, eds.), Cambridge
University Press, Cambridge, UK, pp. 317.

O' Riordan, T., Voisey, H., 1998, The political economy of the sustainability transition, in:
The transition to sustainability: the politics of Agenda 21 in Europe (T. O'riordan, H.
Voisey, eds.), London.

O'Riordan, T., Stoll-Kleemann, 2002, Biodiversity, sustainability and human comunities.

Protecting beyond the protected (T. O'riordan, Stool-Kleemann, eds.), Cambridge
University Press, Cambridge, UK, pp. 317.

O'Riordan, T., Voisey, H., 1998, Agenda 21. The transition to sustainability. The politics
of Agenda 21 in Europe (T. O'Riordan, H. Voisey, eds.), Earthscan Publication Ltd,
London, pp. 320.

226
OECD, 2002, Governance for sustainable development. Five OECD case studies (O. f. e. c.
a. OECD, ed.), Organization for economic cooperation anddevelopment OECD, pp. 384.

OECD, 2006, The new rural paradigm. Policies and governance, in: OECD Rural Policy
Reviews (OECD, ed.), OECD Paris.

Okkonen, L., 2008, From exogenous to endogenous development in Scottish forestry: the
feasibility of small-scale wood energy enterprise, Journal of Environmental Planning and
Management 51(2):221-232.

Olivier, C. D., Kimmins, J. P. H., Harshaw, H. W., Sheppard, S. R. J., 2000, Criteria and
Indicators of sustainable forestry: A systems approach, in: Forests and Landscapes-linking
ecology, sustainability and aesthetics (S. R. J. Sheppard, H. Harshaw, eds.), CABI in
association with IUFRO, Wallingford, Oxon, pp. 73-93.

Ostrom, E., 2007, A diagnostic approach for going beyond panaceas, Proceedings of the
National Academy of Sciences of the United States of America 104(39):15181-15187.

Ostrom, E., Janssen, M. A., Anderies, J. M., 2007, Going beyond panaceas, Proceedings of
the National Academy of Sciences of the United States of America 104(39):15176-15178.

Palmer, J. F., Hoffman, R. E., 2001, Rating reliability and representation validity in scenic
landscape assessments, Landscape and Urban Planning 54(1-4):149-161.

Pearce, D., 2001, The economic value of forest ecosystems, Ecosystems health 7(4):284-
296.

Pereira, E., Queiroz, C., Pereira, H. M., Vicente, L., 2005, Ecosystem Services and Human
Well-Being: A participatory study in a mountain community in Portugal, Ecology &
Society 10(2):14.

Pereira, P. M., Fonseca, M. P., 2003, Nature vs nurture:the making of montado ecosystem,
Conservation ecology 7(3):7.

Perz, S. G., 2007, Grand theory and context-specificity in the study of forest dynamics:
forest transition theory and other directions, The Professional Geographer 59(1):105-114.

Peterson, G., Allen, C. R., Holling, C. S., 2008, Ecological resilience, biodiversity and
scale, Ecosystems 1:6-18.

227
Pina, A. M., Aubyn, M. S., 2005, Comparing macroeconomic returns on human and public
capital: An empirical analysis of the Portuguese case (1960-2001), Journal of Policy
Modelling 27:585-598.

Pinto-Correia, T., 2000, Future development in Portuguese rural areas: How to manage
agricultural support for landscape conservation?, Landscape and Urban Planning 50(1-
3):95-106.

Pinto-Correia, T., Breman, B., 2008, New roles for farming in a differentiated countryside:
the Portuguese example, Regional Environmental Change.

Pommerening, A., Murphy, S. T., 2004, A review of the history, definitions and methods
of continuous cover forestry with special attention to afforestation and restocking, Forestry
77(1):27-44.

Potschin, M., Haines-Young, R., 2006a, "Rio+10", sustainability science and Landscape
Ecology, Landscape and Urban Planning 75(3-4):162-174.

Potschin, M. B., Haines-Young, R. H., 2006b, Landscapes and sustainability, Landscape

and Urban Planning 75(3-4):155-161.

Powers, R. F., 1999, On the sustainable productivity of planted forests, New Forests
17:263-306.

Pretty, J., 2003, Social capital and the collective management of resources, Science
302(5652):1912-1914.

Pretty, J., Ward, H., 2001, Social capital and the environment, World Development
29(2):209-227.

Rekola, M., Pouta, E., 2005, Public preferences for uncertain regeneration cuttings: a
contingent valuation experiment involving Finnish private forests, Forest Policy and
Economics 7(4):635-649.

Rescia, A., Astrada, E., Bono, J., Blasco, C., Meli, P., Adamoli, J., 2006, Environmental
analysis in selection of alternative corridors in a long distance linear project: a
methodological proposal, Journal of Environmental Management 80:266-278.

Ribe, R. G., 1982, On the Possibility of Quantifying Scenic Beauty - a Response,

Landscape Planning 9(1):61-74.

228
Ribe, R. G., 1989, The aesthetics of forestry: what has empirical preference research taught
us?, Environmental Management 13(1):55-74.

Ribe, R. G., 2002, Is scenic beauty a proxy for acceptable management? The influence of
environmental attitudes on landscape perceptions, Environment and Behavior 34(6):757-
780.

Roche, M., 1998, Planting power, the afforestation of the commons and state formation in
Portugal, Journal of Rural Studies 14(2):268-269.

Rogge, E., Nevens, F., Gullink, H., 2007, Perception of rural landscapes in
Flandres:Looking beyond aesthetics, Landscape and Urban Planning 82(4):159-173.

Roovers, P., Hermy, M., Gullink, H., 2002, Visitor profile, perceptions and expectations
from a gradient of increasing urbanisation in central Belgium, Landscape and Urban
Planning 59:129-145.

Rudel, T. K., Coomes, O. T., Moran, E., Achard, F., Angelsen, A., Xu, J. C., Lambin, E.,
2005, Forest transitions: towards a global understanding of land use change, Global
Environmental Change-Human and Policy Dimensions 15(1):23-31.

Sabatier, P. A., Focht, W., Lubell, M., Trachtenberg, Z., Vedlitz, A., Matlock, M., 2005,
Swimming Upstream. Collaborative Approaches to Watershed Management (P. A.
Sabatier, W. Focht, M. Lubell, Z. Trachtenberg, A. Vedlitz, M. Matlock, eds.), MIT Press,
Cambridge, Massachusetts, USA.

Sachs, J. D., 2004, Sustainable Development, Science 304(30 April 2004):649.

Santos, F. D., Forbes, K., Moita, R., 2001, Climate change in Portugal: Scenarios, Impacts
and adaptation measures SIAM. Executive summary and conclusions (GRADIVA, ed.),
Lisbon.

Sayer, J. A., Magginnis, S., 2005, Forests in landscapes. Ecosystem approaches to

sustainability, in: The Earthscan forestry library (J. A. S. S. Maginnis, ed.), IUCN,
London, Sterling, VA, pp. 257.

Sayer, J. A., Maginnis, S., 2005, Forests in landscapes. Ecosystem approaches to

sustainability (J. A. Sayer, S. Maginnis, eds.), IUCN

The Earthscan forestry library, London, Sterling, VA, pp. 257.

229
Schmid, S., Thurig, E., Kaufmann, E., Lischke, H., Bugman, H., 2006, Effect of forest
management on future carbon pools and fluxes, Forest Ecology and Management 237:65-
82.

Selman, P., 1997, The role of forestry in meeting planning objectives, Land Use Policy
14(1):55-73.

Selman, P., 2002, Multi-function landscape plans: a missing link in sustainability

planning?, Local Environment 7(3):283-294.

Selman, P., 2006, Planning at the landscape scale, Routledge, London and New York, pp.
213.

Selman, P., Knight, M., 2006, On the nature of virtuous change in cultural landscapes:
Exploring sustainability through qualitative models, Landscape Research 31(3):295-307.

Seymour, F., 2007, Seeking solutions: Our planet the future of forests (FAO, ed.).

Shearer, E. L., 2005, Approaching scenario based studies:three perceptions about the future
and considerations for landscape planning, Environment and Planning B: Planning and
design 32:67-87.

Shelby, B., Thompson, R., Brunson, M., Johnson, R., 2005, A decade of recreation ratings
for six silviculture treatments in Western Oregon, Journal of Environmental Management
75:239-246.

Sheppard, S. R. J., 2005, Participatory decision support for sustainable forest management:
a framework for planning with local comunities at the landscape level in Canada.
Sustainable forestry in the balance, Journal of Forestry 4:187-195.

Sheppard, S. R. J., Harshaw, H. W., 2000, Forests and landscapes:linking ecology,

sustainability and aesthetics (S. R. J. a. H. W. H. e. Sheppard, ed.), CABI publishing,
Wallingford, Oxon.

Sheppard, S. R. J., Harshaw, H. W., 2001, Priorities for reconciling sustainability and
aesthetics in forest landscape management, in: Forests and lansdcapes-linking ecology,
sustainability and aesthetics (S. Sheppard, H. Harshaw, eds.), CABI publisher in
association with IUFRO, pp. 294.

Sheppard, S. R. J., Harshaw, H. W., McBride, J. R., 2000, Priorities for reconciling
sustainability and aesthetics in forest landscape management, in: Forests and lansdcapes-

230
linking ecology, sustainability and aesthetics (S. Sheppard, H. Harshaw, eds.), CABI
publisher in association with IUFRO, Wallingford, Oxon, pp. 263-287.

Shifley, S., 2006, Sustainable forestry in the balance, Journal of Forestry 4,:187-195.

Silveira, L., 2000, Origins and evolution of the Portuguese administrative system in
comparative perspective (U. N. d. Lisboa, ed.), Universidade Nova de Lisboa, Lisbon.

Silvennoinen, H., Alho, J., Kolehmainen, O., Pukkala, T., 2001, Prediction models of
landscape preferences at forest stand level, Landscape and Urban Planning 56:11-20.

Siry, J. P., Cubbage, F. W., Ahmed, M. R., 2005, Sustainable forestry management: global
trends and opportunities, Forest Policy and Economics 7:551-561.

Slee, B., 2006, The socio-economic evaluation of the impact of forestry on rural
development: A regional level analysis, Forest Policy and Economics 8(5):542-554.

Slee, B., 2007a, Landscape goods and services related to forestry land use, in:
Multifunctional land use: Meeting future demands for landscape goods and services (U.
Mander, H. Wiggering, K. Helming, eds.), Springer, Berlin, pp. 64-82.

Slee, B., 2007b, Social indicators of multifunctional rural land use: The case of forestry in
the UK, Agriculture Ecosystems & Environment 120(1):31-40.

Slee, B., Roberts, D., Evans, R., 2004, Forestry in the rural economy: a new approach to
assessing the impact of forestry on rural development, Forestry 77(5):441-453.

Slee, B., Snowdon, P., 1999, Rural development forestry in the United Kingdom, Forestry
72(3):273-284.

Southern, A., 2008, Implementing an integrated approach to natural resource governance:

A case study in whole landscape design at the catchment scale. PhD Thesis, in: School of
Environmental Sciences, University of East Anglia, Norwich, pp. 230.

Spiecker, H., 2003, Silvicultural management in mantainning biodiversity and resistence of

forests in Europe, Journal of Environmental Management 67:55-65.

Stengera, A., Haroub, P., Navrudc, S., 2009, Valuing environmental goods and services
derived from the forests, Journal of Forest Economics 15(1-2):1-14.

Stevens, J. A., Montegomerey, C., 2002, Understanding the compatibility of multiple uses
on forest land: A survey of multiresource research with application to the Pacific

231
Northwest, in: General technical report PNW-GTR-539, United States Department of
Agriculture (USDA, ed.), Washington.

Steyaert, P., Ollivier, G., 2007, The European Water Framework Directive: How
ecological assumptions frame technical and social change, Ecology and Society 12(1):art
25, Online.

Strassburg, B., Turner, K., Fisher, B., Schaeffer, R., Lovett, A., 2009, Reducing emissions
from deforestation- The "combined incentives" mechanism and empirical simulations,
Global Environmental Change in press.

Surova, D., Pinto-Correia, T., 2008, Landscape preferences in the cork oak Montado
region of Alentejo, southern Portugal: Searching for valuable landscape characteristics for
different user groups, Landscape Research 33(3):311-330.

Tahvanainen, L., Tyrvainen, L., Ihalainen, M., Vuorela, N., Kolehmainen, O., 2001, Forest
management and public perceptions - visual versus verbal information, Landscape and
Urban Planning 53(1-4):53-70.

Thorne, J. F., Huang, C. S., 1991, Toward a landscape ecological aesthetic: methodologies
for designers and planners, Landscape and Urban Planning 21:61-79.

Tips, W. E. J., Vasdisara, T., 1986, The influence of the socio-economic background of
subjects on their landscape preference evaluation, Landscape and Urban Planning
13(Short communication):225-230.

Tovey, H., 2008, Introduction:Rural Sustainable Development in the Knowledge Society

Era, Sociologia Ruralis 48(3):185-199.

Tress, B., Tress, G., 2001, Capitalising on multiplicity: a transdisciplinary systems

approach to landscape research, Landscape and Urban Planning 57(3-4):143-157.

Tress, B., Tress, G., 2003, Scenario visualisation for participatory landscape planning - a
study from Denmark, Landscape and Urban Planning 64(3):161-178.

Tress, B., Tress, G., Decamps, H., d'Hauteserre, A. M., 2001, Bridging human and natural
sciences in landscape research, Landscape and Urban Planning 57(3-4):137-141.

Turner, K. R., Daily, G. C., 2008, The Ecosystem Services Framework and Natural Capital
Conservation, Environmental and Resource Economics 39:25-35.

232
Vejre, H., Abildtrup, J., Andersen, E., Anerden, P., Brandt, J., Busck, A., Dalgaard, T.,
Hasler, B., Huusom, H., Kristensen, L., Kristensen, S., Praestholm, S., 2007,
Multifunctional agriculture and multifunctional landscapes- land use as an interface, in:
Multifunctional land use. Meeting future demands for landscape goods and services (U.
Mander, H. Wiggering, K. Helming, eds.), Springer, Berlin, pp. 93-104.

von Haaren, C., Warren-Krestzchmar, B., 2006, The interactive landscape plan: use and
benefits of new technologies in landscape planning and discussion of the interactive
landscape plan in Koenigslutter am Elm, Germany, Landscape Research 31(1):83-85.

von Haaren, C. V., Ott, C. G. S., 2008, Landscape planning. The basis of sustainable
landscape development, Federal Agency for Nature Conservation, Leipzig, Germany.

Vora, R., 1997, Developing programs to monitor ecosystems health and effectiveness of
management practices on lake states national forests, USA, Biological Conservation
80:289-302.

Wang, X., Song, B., Chen, J., Crow, T. R., LaCroix, J. J., 2006, Challenges in Visualizing
Forests and Landscapes, Journal of Forestry September.

WCED, 1987, Our common future, Oxford University Press, Oxford and New York, pp.
383.

Winter, C., 2005, Preferences and values for forests and wetlands: A comparison of
farmers, environmentalists, and the general public in Australia, Society & Natural
Resources 18(6):541-555.

WRI, 2009, Forest ecosystems: Forest extent and change (W. R. Institute, ed.),
Washington, USA.

Wu, J. G., 2006, Landscape ecology, cross-disciplinarity and sustainability science,

Landscape Ecology 21(1):1-4.

WWF, 2000, The forest industry in the 21st Century, in: Executive summary (W. Forests,
ed.), Surrey, UK.

Zander, P., Karpinski, I., Knierim, A., 2005, Indicators and methods to measure and assess
multifunctionality: Multiagri project.

Zube, E. H., 1986, Local and extra-local perceptions of National Parks and protected areas,
Landscape and Urban Planning 13:11-17.

233
Appendices

234