decision support tools for forest landscape …decision support tools for forest landscape...

O C C A S I O N A L P A P E R

Robin L Chazdon

Manuel R Guariguata

Decision support tools for forest landscape restorationCurrent status and future outlook

Decision support tools for forest landscape restorationCurrent status and future outlook

Robin L ChazdonTropical Forests and People Research Centre, University of the Sunshine CoastInternational Institute of Sustainability

Manuel R GuariguataCIFOR

OCCASIONAL PAPER 183

Center for International Forestry Research (CIFOR)

Occasional Paper 183

© 2018 Center for International Forestry Research

Content in this publication is licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0), http://creativecommons.org/licenses/by/4.0/

ISBN 978-602-387-070-7DOI: 10.17528/cifor/006792

Chazdon RL and Guariguata MR. 2018. Decision support tools for forest landscape restoration: Current status and future outlook. Occasional Paper 183. Bogor, Indonesia: CIFOR.

Photo by Tomas Bertelsen/IPELand use mosaic in Morro do Diabo State Park, Sao Paulo State, Brazil.

CIFORJl. CIFOR, Situ GedeBogor Barat 16115Indonesia

T +62 (251) 8622-622F +62 (251) 8622-100E [email protected]

cifor.org

We would like to thank all funding partners who supported this research through their contributions to the CGIAR Fund. For a full list of the ‘CGIAR Fund’ funding partners please see: http://www.cgiar.org/who-we-are/cgiar-fund/fund-donors-2/

Any views expressed in this publication are those of the authors. They do not necessarily represent the views of CIFOR, the editors, the authors’ institutions, the financial sponsors or the reviewers.

iii

Contents

Abbreviations v

Acknowledgements vi

Executive summary vii

1 Introduction 11.1 The challenge of forest landscape restoration 11.2 What are decision support tools? 41.3 Guiding principles of FLR 51.4 Why decision support tools are needed to guide the FLR process 51.5 Scope and organization of this report 7

2 Tools for preparation and assessment 92.1 A diagnostic tool for identifying readiness for FLR 92.2 A tool for assessing restoration needs and ecosystem conditions 102.3 A tool for mapping, monitoring and planning at the landscape scale 122.4 A tool for assessing FLR opportunities and approaches at national or sub-national levels 122.5 Tools for seeking financing for FLR 182.6 Tools for engaging stakeholders in the FLR process 192.7 Concluding remarks 20

3 Tools to evaluate potential restoration outcomes 213.1 Tools and models for assessing and mapping ecosystem service supply 213.2 Tools for modeling synergies and trade-offs between ecosystem services 243.3 Tools for incorporating biodiversity into systematic restoration planning 263.4 Assessing and mapping restoration costs and cost-effectiveness 283.5 Concluding remarks 30

4 Tools for prioritization, spatial planning and species selection 314.1 Systematic restoration planning 314.2 Spatial prioritization tools and their application to restoration 324.3 Spatial tools for prioritizing areas for specific restoration interventions 404.4 Spatial tools for selecting appropriate species for agroforestry and restoration 404.5 Concluding remarks 43

5 Conclusion: Synthesis and outlook 445.1 Tools for what and for whom? 445.2 The power of visualization 445.3 Theory of change for FLR 465.4 New frontiers for decision support tools for FLR 465.5 Challenges to operationalizing FLR 49

References 50

iv

List of figures, tables and boxes

Figures1 Conceptual illustration of the long-term implementation of restoration. 32 Roadmap of this review. 83 The eight steps of planning and monitoring landscape interventions, in which

maps are used within the five-phase adaptive collaborative management process. 124 The phases of the Restoration Opportunity Assessment Methodology. 155 A carbon accrual curve for Rwanda based on their ROAM assessment. 176 Map of restoration opportunities and associated planned interventions from

the Guatemala ROAM assessment. 187 The types and rates of delivery of ecosystem services provided by natural,

semi-natural, planted forest and planted trees outside the forests. 238 Efficiency frontiers with optimal restoration allocation maps. 269 Priority sites for restoration in Mexico based on biological importance and

restoration feasibility criteria. 2810 A preliminary analysis of priority areas identified through multi-criteria analysis. 3311 Maps of the potential for restoration in Uganda to increase carbon storage (A),

improve water quality (B), increase biodiversity (C) and incur opportunity costs (D). 3912 Maps of priority areas for agroforestry restoration under four social-ecological

scenarios. 4113 A flowchart showing user steps when using a decision support tool for tree selection

on smallholders’ coffee farms. 4314 Smart-maps created by villagers in East Sudan. 4515 A proposed framework for participatory restoration in rural areas of North Africa. 48

Tables1 International organizations, multilateral conventions, and networks with active forest

and landscape-scale restoration programs. 22 The ten principles of the landscape approach. 73 Results of the restoration diagnostic in Rwanda. 114 Four key questions to provide information needed to decide on the optimal

tree cover of a landscape 165 Decision support tools for assessment and mapping of ecosystem services. 226 Components of ecological restoration costs and factors that can be used to estimate

and model these costs. 297 Spatial prioritization studies for restoration planning. 35

Boxes1 Guiding principles of forest landscape restoration 42 Freely available support tools discussed in this chapter 14

v

Abbreviations

ARIES Artificial Intelligence for Ecosystem Services

FAO Food and Agriculture Organization of the United Nations

FLAT Forest Landscape Assessment Tool

FLR Forest landscape restoration

FPIC Free, Prior and Informed Consent

GDP Gross domestic product

InVEST Integrated Valuation of Ecosystem Services and Trade-offs

IUCN International Union for the Conservation of Nature

Ilm Integrated landscape management

LOAM Landscape Outcome Assessment Methodology

MIMES Multiscale Integrated Models of Ecosystem Services

MEA Millennium Ecosystem Assessment

NCP Nature’s Contributions to People

PNV Potential natural vegetation

REDD Reducing Emissions from Deforestation and Forest Degradation

ROAM Restoration Opportunities Assessment Methodology

ROI Return on investment

ROOT Restoration Opportunities Optimization Tool

SLM Sustainable land management

SolVES Social Values for Ecosystem Services Tool

TEEB The Economics of Ecosystems and Biodiversity

TESSA Toolkit for Ecosystem Service Site-based Assessment

TNC The Nature Conservancy

UN United Nations

UNCBD United Nations Convention on Biological Diversity

UNCCD United Nations Convention to Combat Desertification

UNFCCC United Nations Framework to Combat Climate Change

UNEP United Nations Environment Programme

WWF World Wildlife Fund

WRI World Resources Institute

vi

Acknowledgements

We thank Bruno Locatelli for reviewing an initial draft. Also, thanks are due to Erisa, Gideon Suharyanto and Ina Rachmawati for logistical support during the production phase. This work was funded by the CGIAR Program on Forests, Trees and Agroforestry and the United States Agency for International Development (USAID).

vii

Executive summary

Decision-making bodies at all scales face an urgent need to conserve remaining forests and reestablish forest cover in deforested and degraded forest landscapes. The scale of the need, and the opportunity to make a difference, is enormous. Degradation is often viewed as ‘the problem’ and restoration as ‘the solution’. But, rather than being a goal, restoration is the means to achieve many goals. Forest landscape restoration (FLR) is an active, long-term process to regain ecological integrity and enhance human well-being in landscapes that have lost forest cover, forest qualities and forest-based contributions to people.

FLR provides a framework for implementing restoration interventions that collectively address major environmental challenges, such as land degradation, biodiversity loss, water scarcity, lack of sustainable rural livelihoods, and climate change mitigation and adaptation. Restoration interventions can take many forms that vary in cost, trajectory, and specific economic and social outcomes; likewise, their benefits accrue to different actors and stakeholders. These interventions will have a better chance of achieving positive long-term outcomes if the FLR process incorporates clear objectives, good governance, advance preparation and spatial planning.

Many tools, including frameworks, methods, guides, techniques, procedures and analytic approaches, have been developed to guide and inform decision making in the practice of conservation and restoration efforts like FLR. Decision support tools help communities, organizations, and regional and national government agencies, guiding and supporting them with readiness, assessments, mapping specific interventions, revealing trade-offs and synergies of specific approaches, assessing optimal solutions based on spatial or non-spatial criteria, or modeling scenarios based on specific assumptions and data. They focus on biophysical and economic dimensions of resource management, conservation and restoration, but can also provide guidance in social decision making, governance and negotiation among stakeholders. This guidance is

critical, since such multi-sectorial planning and stakeholder engagement are essential conditions for the implementation of successful restoration actions at local and national scales, and for mobilization of adequate financial and public support.

This document provides a comprehensive review of the tools available to guide preparation and decision making, both prior to and during the implementation process. We also address the need for additional tools and analytic approaches. Examples of reviewed tools include: Restoration Diagnostic, a tool for developing national or regional strategies for successful FLR, based on the status of success factors important to motivate, enable and implement restoration interventions; and Restoration Opportunities Assessment Methodology (ROAM), which guides national or sub-national stakeholder groups through a process to become engaged in stock-taking, mapping, analysis of costs and benefits of restoration activities, and assessment of financing and investment options. Our review reveals that decision support tools to guide FLR assessment, capacity building and implementation at landscape scales are yet to be well developed.

The review also describes and compares tools and models developed to assess the supply of ecosystem services. These tools are actively proliferating and evolving. A common approach for assessing the impact of restoration on ecosystem services uses mechanistic process models that predict how multiple ecological variables influence ecosystem service supply, based on land-cover change scenarios. Efficiency frontiers (production possibility frontiers) are a widely used tool for visualizing trade-offs and synergies in ecosystem service supply and biodiversity outcomes during restoration.

Our review demonstrates how tools for biodiversity assessment can play an integral role in FLR planning at regional and landscape scales, including prioritizing locations and types of biological

viii

corridors, assessing species complementarity networks and evaluating potential outcomes of different restoration interventions, on local- and landscape-scale species and functional diversity. Species distribution modeling and habitat suitability modeling can provide decision support for prioritizing areas within landscapes where restoration will likely bring high biodiversity benefits at reduced cost.

Estimates of the economic value of ecosystem services can be generated from ecosystem service production functions, along with benefit transfer for economic valuation and ecosystem service demand functions. Economic analysis and scenarios based on ecosystem service supply and biodiversity benefits can generate projections of cost-effectiveness and total costs of specific restoration interventions in particular areas. As these approaches are limited and case specific, there is a need to develop standardized tools and databases to apply economic valuation methods and to help restoration practitioners develop economic scenarios for particular regions or landscapes to guide decision making and predict financing needs. Socioeconomic outcomes of restoration interventions, such as livelihoods and food security, are also poorly addressed by available decision support tools.

Our review emphasizes that tools for estimating the costs of restoration and the cost-effectiveness of restoration interventions are central to planning and prioritization. In current models, land sale, or rental cost, or gross economic rents for agricultural land, are often used as proxies for opportunity costs. The potential for natural regeneration is commonly used as a proxy for implementation costs; spontaneous forest establishment is far less expensive than full-scale site preparation and tree planting.

We encountered a wide range of studies that apply spatial prioritization models to identify priority areas for specific FLR interventions within regions or countries. This systematic approach is based on spatially explicit social, biophysical, and political data. The main objective of spatial prioritization is to select sites or larger areas for investments and interventions that (1) maximize the long-term success of restoration; (2) ensure representation of biological, geographic and human diversity; and (3) maximize socioecological benefits for local communities and other stakeholders. Given limited resources, the best returns on investments combine feasibility (actions that are practical and possible) with high levels of multiple benefits. Different types of spatial data sets can provide indicators or proxies for feasibility and restoration

benefits. For example, areas with low land cost or high local engagement and commitment indicate a higher feasibility for restoration. Restoration benefits can be assessed and mapped based on the local potential for habitat improvement, landscape connectivity, livelihoods and farm income, ecological resilience, or reduced risk of species loss.

Tools that apply a systematic approach to restoration can identify potential land-use conflicts, evaluate trade-offs among benefits and assess divergent stakeholder needs. Systematic approaches can incorporate a wide range of social, political, economic and ecological dimensions, and are now being widely adapted to identify restoration priority areas, using a variety of software and mapping tools. Recently developed tools provide information about tree and shrub species that have been shown to be useful to farming or pastoral communities throughout eastern Africa and allow users to identify which tree species would likely grow well at particular restoration sites based on their expected ability to persist there under current and future climatic conditions. These tools help to avoid costly mistakes by planting species that are best suited for particular conditions, or by restricting tree-planting interventions to areas where assisted or spontaneous natural regeneration are not feasible outcomes.

Decision support tools are ineffective without a clear objective in mind and without a framework for action. Relatively simple tools are more likely to be adopted by policy makers, provided they are clearly documented, published and have been validated to reveal their limitations. Visualization exercises can enhance communication, stakeholder engagement and encourage constructive dialogue with policy makers and community leaders regarding restoration scenarios.

Despite the many advances in the development and application of decision support tools in FLR, this review reveals a gap in tools for the implementation of landscape-scale restoration initiatives and for guiding monitoring and adaptive management. The review also reveals that available tools primarily focus on assessing restoration opportunities at a broader scale (within countries or regions), rather than within landscapes where implementation occurs. Evidence from research on community-based conservation and forest management suggests that tools for the empowerment, land rights and capacity building of local residents can help nurture strong coalitions of landscape restoration practitioners that apply adaptive management of restoration interventions, and evaluate potential restoration scenarios in their own landscapes.

1 Introduction

1.1 The challenge of forest landscape restoration

We live in a world of diminishing forests, reduced supply of goods and services that forests provide to people, declining forest-dependent biodiversity and uncertain climate change. Decision-making bodies at all scales face an urgent need to conserve remaining forests, and reestablish forest cover in deforested and highly altered forest areas. The scale of the need, and the opportunity to make a difference, is enormous.

Across the tropics, most opportunities for restoration in forest biomes lie within mixed-use landscapes. One influential global assessment showed that over 2 billion ha of deforested and degraded landscapes could be considered as opportunities for restoration interventions (Laestadius et al. 2012). A global-scale map has shortcomings, but nevertheless provides a critically important tool for motivating countries, organizations and policy makers around the world to ‘see’ the enormous potential for restoration within their jurisdictions. It invites decision makers to take a closer look at where and how particular restoration interventions could be planned and implemented in particular sites within local areas of interest and authority.

Forest landscape restoration (FLR) provides a framework for implementing restoration interventions that collectively address major environmental challenges, such as land degradation, biodiversity loss, water scarcity, lack of sustainable rural livelihoods, and climate change mitigation and adaptation. Restoration will be most likely to achieve these long-term outcomes if the process incorporates clear objectives, good governance, advance preparation, and spatial planning. In many ways, the stages involved in preparing for large-scale restoration are similar to the readiness activities undertaken in the 64 developing countries

involved in the United Nations Program on Reducing Emissions from Deforestation and Forest Degradation (REDD) (Cerbu et al. 2011).

At the national level, the process of achieving restoration begins long before implementation. Recognizing both the need and potential for restoration is the first step to motivate action, followed by a formal commitment to implement restoration interventions. To this end, multiple international and global initiatives have arisen to encourage national-scale commitments to undertake restoration (Table 1).

Although these commitments signal the first steps of a complex and long-term restoration trajectory, the next steps of planning, prioritizing and implementing restoration interventions require overcoming many obstacles and knowledge gaps (Chazdon et al. 2017; Mansourian et al. 2017). These steps involve a series of planning, capacity building and stakeholder engagement activities that culminate in a restoration and monitoring plan at the national or sub-national scale. These planning steps require tools to guide decision making, extensive stakeholder engagement, and the application of the best local and expert knowledge available. Restoration is an iterative process, involving regular review of design and implementation stages through monitoring and evaluation, with adjustment as necessary to ensure that original goals are met and that the restoration process retains its core principles of being effective, efficient and engaging (Keenleyside 2012; Figure 1). Implementation involves seven key steps repeated over time, using adaptive management approaches.

Numerous case studies illustrate the benefits of planning, capacity building and broad stakeholder engagement to ensure successful and long-lasting restoration activities (Vallauri et al. 2005; Le et al. 2014; Mansourian and Vallauri 2014; Reid et al. 2017). Many restoration and reforestation efforts,

| Robin L Chazdon and Manuel R Guariguata2

Table 1. International organizations, multilateral conventions, and networks with active forest and landscape-scale restoration programs.

Organization/network Program

Global Partnership on Forest and Landscape Restoration (GPFLR)

Learning network/discussion platform on forest landscape restoration

International Union for the Conservation of Nature (IUCN) (Global Forest and Climate Change Program)

Forest landscape restoration initiative

World Resources Institute (WRI) Global restoration initiative, 20x20 initiative (Latin America)

New Partnership for Africa’s Development (NEPAD) AFR100 initiative (the African forest landscape restoration initiative)

United Nations Food and Agriculture Organization (FAO)

Forest and landscape restoration mechanism

United Nations Convention on Biological Diversity (CBD)

Forest ecosystem restoration initiative

Collaborative Group on International Agricultural Research (CGIAR)

Research program on Forests, Trees, and Agroforestry

People and Reforestation in the Tropics, a Network for Education, Research, and Synthesis (PARTNERS)

Synthetic research, policy outreach, educational outreach

African Union The Great Green Wall for the Sahara and the Sahel initiative

United Nations Environment Program (UNEP) One Billion Tree program

The Nature Conservancy (TNC) Plant a Billion Trees

World Wildlife Fund (WWF) Landscape restoration in 10 ecoregions

Ecological Restoration Alliance of Botanical Gardens 100 restoration projects on several continents, over next 20 years

EcoAgriculture Partners Landscapes for Food, People, and Nature

Natural Resources Canada—Canadian Forest Service (NRCan-CFS)

International ‘model forest’ network

Source: adapted from Chazdon et al. (2017).

of all sizes, have failed. High seedling mortality characterizes many restoration projects, due to the planting of species that are not adapted to local conditions, poor genetic quality of planting stock, and poor nursery or planting practices (Gregorio et al. 2016). In other cases, large-scale monoculture tree plantations became susceptible to disease, resulting in massive tree mortality (Abbas et al. 2016). Restoration plantings are frequently destroyed by fires or cleared for agricultural use, particularly in cases where local communities are not actively engaged in planning or implementation (Chokkalingam et al. 2006). Poor planning and corruption are also associated with the failure of reforestation projects (Le et al. 2012). Despite 30 years of government-managed forest rehabilitation and restoration programs in Indonesia, the area of degraded forest has doubled since the 1970s, while rehabilitation efforts have largely failed (Nawir et al. 2007; Barr and Sayer 2012). The Program

on Reforestation and Afforestation initiated by presidential decree in Indonesia in the 1970s received 85% of the Indonesian government’s total forestry budget, but showed little to no long-term success (Nawir et al. 2007).

Achieving success in restoration initiatives requires consideration of short- and long-term economic and social outcomes for local communities, through their active engagement in the process (Kang and Kim 2015). To be successful in the long term, restoration initiatives must address the underlying causes of deforestation and degradation; they also need to be proactive, with the participation of multiple stakeholders, and long-term planning horizons (Nawir et al. 2007; Le et al. 2012). Restoration initiatives need to move beyond time- and space-bound project-based activities to foster real change in the management and governance of forest landscapes.

Decision support tools for forest landscape restoration | 3

Restoring forests is a long-term socioecological process that requires several human generations to achieve its full trajectory (Chazdon et al. 2017). Forest restoration is strongly aligned with policy objectives embodied by the UN’s Sustainable Development Goals, including increasing food security, reducing poverty, enhancing water supplies, protecting biodiversity, disaster mitigation and climate-change adaptation (Mansourian and Vallauri 2014; Stanturf et al. 2015). A complex set of biophysical and social factors need to be considered in making decisions regarding where and how to restore forests. This includes geology, soils and topography; land-use history; landscape composition and connectivity; conservation status of species; land tenure, food security, economic needs and livelihoods of local communities; and the site-level potential to restore vital ecosystem functions, such as regulation of water yield, sediment retention and carbon storage. Most important is that the objectives of restoration interventions and their spatial locations are aligned with stakeholder needs, biophysical conditions, government policies, and institutional capacities and potential to adapt.

Stanturf et al. (2017) break down the FLR process into four distinct phases. The first phase is a visioning phase, where FLR is placed within a specific context in a country, region, or landscape. Visioning requires knowing where degradation or forest loss has occurred, and what actions and interventions constitute FLR (Box 1). The second is a conceptualizing phase, where priority landscapes and specific ecological and social goals are identified. This phase requires assessments of restoration opportunities, using biophysical and social criteria. Multiple stakeholder groups participate in these assessments, based on specific objectives for different FLR interventions (Mansourian and Vallauri 2014). The third phase involves designing the FLR project, based on the objectives selected in the second phase, with clearly identified start and end points for restoring specific landscape elements. The fourth is the implementation phase, where detailed plans for interventions are put into place to meet the objectives for FLR. Monitoring is a critical element of implementation. Implementation decisions involve selecting specific sites for applying restoration interventions, selecting species for planting, planting designs and genetic provenances.

Figure 1. Conceptual illustration of the long-term implementation of restoration.

Source: Keenleyside (2012, Figure 4).


1.2 What are decision support tools?

Decision support tools include frameworks, methods, guides, techniques, procedures and analytic approaches that are targeted to a specific group to inform decision making (De Ridder et al. 2007). Tools can be as simple as a flowchart, or as complex as a packaged system of computer models. They help to operationalize principles and concepts in practice. For example, a compass is a hand-held tool for drawing perfect circles. Without a compass it is easy to conceptualize a perfect circle, but difficult to draw one. Decision support

tools are developed with the goal of guiding or supporting readiness, assessment, mapping specific interventions, revealing trade-offs and synergies of specific approaches, assessing optimal solutions based on spatial or non-spatial criteria, or modeling scenarios based on specific assumptions and data. Tools (also called methods, instruments, frameworks or models) are tailored for particular contexts and utilize information fit to purpose.

Many tools have been developed to provide guidance with decision making in the practice of conservation and restoration. Some tools are more focused on

Box 1. Guiding principles of forest landscape restoration

• Focus on landscapes Implement restoration interventions within a broad landscape context. Landscapes are mosaics of interacting land uses and land holdings, including managed and unmanaged ecosystems.

• Engage stakeholders and support participatory governance Actively engage local stakeholders, including particularly vulnerable groups, in planning and decision making regarding restoration goals and strategies, implementation methods, resolution of land-use and tenure issues, governance structures, benefit sharing, monitoring and review processes.

• Restore ecological functions Restore ecological functions across a landscape by improving the quality of habitats for wildlife and the diversity of species, enhancing productive land uses, preventing erosion and flooding, and increasing resilience to climate change and other disturbances. Integrate forest functions into land-use management plans supporting conservation and sustainable use of forests.

• Use a variety of restoration approaches Consider a wide range of approaches in developing restoration plans for implementation within the landscape combining the technical strategies that are most suitable, including natural regeneration, agroforestry and different types of tree plantings, implemented in appropriate landscape settings.

• Conserve and enhance natural ecosystems Stop further deforestation and degradation of natural forests and other associated ecosystems, and enhance the recovery and conservation of forest remnants within the landscape.

• Tailor to local context. Adapt restoration approaches to the local social, cultural, economic and ecological values and needs.

• Restore the provision of a diversity of goods and services Generate a suite of ecosystem goods and services that benefit multiple stakeholder groups. Goods and services thus generated could improve soil fertility, reduce erosion, provide shade, produce timber and non-timber forest products, store carbon, increase downstream water supply and quality, and improve wildlife habitats.

• Manage adaptively over the medium and long-term Adjust restoration approaches over time as environmental conditions, knowledge, capacity, stakeholder needs, and societal values change. Integrate information and knowledge from monitoring activities, research and development, and stakeholder guidance into management plans and decisions as restoration progresses from initiation to improved ecological function.

Source: based on WRI (2015) and Maginnis and Jackson (2005).


ways that balance multiple social and environmental outcomes and benefit multiple stakeholders. Within the context of FLR, restoration interventions can take on many forms and spatial extents. Under favorable social and ecological conditions, interventions can take the form of spontaneous, assisted or farmer-managed natural regeneration (Chazdon and Guariguata 2016). Alternatively, active restoration interventions can involve extensive planting of native or exotic nursery-grown tree seedling in species mixtures. In some cases, topsoil amendments may be needed to reestablish forest cover after mining or bulldozing activity. Agroforestry systems, such as silvopastoral systems, alley-cropping systems involving crops growth with timber or fruit trees, or shade-grown coffee or cacao beneath forest canopies, can also serve as restoration interventions. Forestry plantings, such as woodlots and commercial tree plantations, can also be components of FLR, provided they satisfy the principles (Box 1). Each of these interventions has associated implementation and opportunity costs, as well as economic and social outcomes, and benefits that accrue to different actors and stakeholders. Additionally, these interventions can vary in their spatial extent and proximity to other land uses which also influence landscape functions and the supply of goods and services.

Who will make decisions on restoration interventions, and how will they be made? Decision support tools can help communities, organizations, and regional and national government agencies to identify what type and combinations of interventions are most appropriate, and most cost-effective, for different spatial locations and land-use contexts. In many respects, FLR is a specialized form of integrated landscape management (ILM) or sustainable land management (SLM), approaches which are being applied around the world to increase food security, ecosystem services, biodiversity and climate resilience in agricultural landscapes (Sayer et al. 2013; Reed et al. 2016). Decisions are ultimately implemented within landscapes, and therefore must engage stakeholder groups at this level.

1.4 Why decision support tools are needed to guide the FLR process

Principles, guidelines and best practices have been developed for ecological restoration in protected areas (Keenleyside et al. 2012) and international standards for ecological restoration were recently

steering process-oriented decisions and assessing readiness for particular types of restoration actions, whereas others are used to perform quantitative assessments and develop scenarios for restoration outcomes, given a set of data layers. Decision support tools focus on biophysical and economic dimensions of resource management, conservation and restoration, but can also provide guidance in social decision making, governance and negotiation among stakeholders (Lynam et al. 2007; Reed 2008; van Oosten et al. 2014).

1.3 Guiding principles of FLR

Today, most policy makers embrace FLR as their approach to large-scale restoration, as it encompasses both social and environmental dimensions, and can flexibly accommodate trade-offs among different types of activities within landscapes (Mansourian et al. 2017). For this reason, regional and global restoration initiatives advocate FLR practice (Table 1). The FLR approach originated in 2000, in recognition of the limitations of small-scale ecological restoration and large-scale monoculture forestry to achieve a desirable balance between meeting the needs of people, and restoring functionality within landscapes (Maginnis and Jackson 2005; Mansourian et al. 2005; Laestadius et al. 2015).

The decision support tools reviewed here are based on the FLR model and principles (Box 1). The original definition of FLR is “a planned process that aims to regain ecological integrity and enhance human well-being in deforested or degraded landscapes” (WWF and IUCN 2000, 2). More recently, the Global Partnership on Forest Landscape Restoration has defined FLR as “an active process that brings people together to identify, negotiate and implement practices that restore an agreed optimal balance of the ecological, social and economic benefits of forests and trees within a broader pattern of land uses” (Sabogal et al. 2015, 4). Depending on the circumstances, FLR can benefit from centralized planning, financial backing and technical support, in cooperation with local communities and institutions, or can achieve successful outcomes within landscapes in the absence of a formal planning process.

Regardless of the role of centralized planning, FLR requires key decisions to be made regarding where and how to restore forests in landscapes in


proposed (McDonald et al. 2016). These considerations are certainly important for restoration of forests within the broader context of FLR. But FLR does not simply focus on planting trees or restoring forest ecosystems (Mansourian et al. 2005), and the complexity of planning and implementing FLR reaches beyond a focus on particular ecosystems (Chazdon 2017). The broad objectives of FLR are rooted in a landscape approach, which is defined as ‘‘a long-term collaborative process bringing together diverse stakeholders aiming to achieve a balance between multiple and sometimes conflicting objectives in a landscape or seascape’’ (Sayer et al. 2017, 465).

The landscape approach (Table 2) offers a promising framework for FLR planning and implementation, but an evidence base is needed to illustrate the social and environmental benefits (Reed et al. 2016, 2017). To achieve a balance among the many social and environmental objectives of FLR, the juxtaposition of multiple layers of spatially explicit information can provide important guidance for spatial planning and prioritization. In addition, the objectives and approaches taken in planning and implementation must be linked to the local context, synthesizing local knowledge with broader objectives, such as national or sub-national policies. Spatial planning can help to reduce overall costs and lead to the equitable delivery of benefits to multiple stakeholders.

There is no single blueprint for how to restore a landscape. For example, in an upland region, restoration efforts may aim to protect watersheds through active tree planting and agroforestry plantations on degraded land. In a lowland region, restoration efforts may be focused on intensifying agroforestry systems and establishing mixed-species silvicultural plantations. In landscapes with remnant forest fragments, restoration efforts might focus on creating biological corridors, and encouraging natural regeneration in buffer zones surrounding fragments to protect threatened biodiversity.

The key to successful restoration lies in achieving the desired impacts, which are a product of the initial vision, planning, implementation, monitoring and adaptive management. Planning and prioritization activities are essential to reach scale and to allocate limited financial resources

to maximize impact. All of these activities require active stakeholder engagement to ensure that they meet the needs of local communities and have the full support of local governments and organizations. Given that funds to support restoration activities are limited, cost-effectiveness is an essential criterion in planning and implementation: how can the greatest impacts be achieved for a limited level of investment?

FLR planning and implementation tools share some functionality with tools developed for conservation, such as Marxan, which identifies the optimal location of protected areas with biodiversity representation goals (Ball et al. 2009). But some new functions are required to address the complexity of issues, and to help identify the best match between restoration interventions, needs of local communities, types of degradation, and national priorities (see Chapter 3). Decision support tools can help to identify important trade-offs and synergies between different restoration objectives and potential outcomes, as well as to evaluate the cost-effectiveness of restoration interventions at different spatial scales. Decision support tools can be used to prioritize regions and specific sites for restoration, based on single or multiple criteria, including provision of particular types of ecosystem services.

Embarking on a successful restoration path requires recognizing failures of past reforestation and restoration efforts, and elucidating the conditions that define successful restoration outcomes within a local context (Nawir et al. 2007; Le et al. 2014). Assessment of key success factors prior to restoration helps to ensure that the necessary elements for planning and implementation are in place (WRI 2015; Figure 1). Both successes and failures are context-dependent, and should be evaluated in terms of clearly stated objectives that form the foundation of planning and implementation strategies. To successfully reverse long-term patterns of degradation, poverty, and food insecurity, FLR needs to become the new status quo for sustainably managing productive and non-productive land in forest landscapes.

Restoration is a means to an end, rather than a goal in itself. Whether or not restoration interventions are successful in meeting their objectives will depend heavily on realistic time horizons, adequate financial investment,


Table 2. The ten principles of the landscape approach.

Principle Description

Principle 1: Continual learning and adaptive management

Landscape processes are dynamic. Despite the underlying uncertainties in causes and effects, changes in landscape attributes must inform decision making. Learning from outcomes can improve management.

Principle 2: Common concern entry point

Identifying immediate ways forward, through addressing simpler short-term objectives, can begin to build trust with stakeholders. Each stakeholder will only join the process if they judge it to be in their interest.

Principle 3: Multiple scales

Numerous system influences and feedbacks affect management outcomes, but these impacts unfold under the influence of a range of diverse external influences and constraints. An awareness of higher and lower level processes can improve local interventions, inform higher-level policy and governance, and help coordinate administrative entities.

Principle 4: Multifunctionality

Landscapes and their components have multiple uses and purposes, each of which is valued in different ways by different stakeholders. Trade-offs exist among the differing landscape uses; these need to be reconciled in a spatially explicit, ecosystem-driven manner, that acknowledges stakeholders’ multiple needs, preferences and aspirations.

Principle 5: Multiple stakeholders

Multiple stakeholders frame and express objectives in different ways. Failure to recognize and engage stakeholders in an equitable manner in decision-making processes will lead to sub-optimal, and sometimes unethical, outcomes.

Principle 6: Negotiated and transparent change logic

Transparency is the basis of trust among stakeholders, and is achieved through a mutually understood and negotiated process of change; helped by good governance. All stakeholders need to understand and accept the general logic, legitimacy, and justification for a course of action, and to be aware of the risks and uncertainties.

Principle 7: Clarification of rights and responsibilities

Rules on resource access and land use shape social and conservation outcomes, and need to be clear, as a basis for good management. Access to a fair justice system allows for conflict resolution and recourse. The rights and responsibilities of different actors need to be clear to, and accepted by, all stakeholders.

Principle 8: Participatory and user-friendly monitoring

To facilitate shared learning, information needs to be widely accessible. Systems that integrate different kinds of information need to be developed. When stakeholders have agreed on desirable actions and outcomes, they will share an interest in assessing progress.

Principle 9: Resilience System-level resilience can be increased through an active recognition of threats and vulnerabilities. Actions need to be promoted that address threats and allow recovery after perturbation, through improving capacity to resist and respond.

Principle 10: Strengthened stakeholder capacity

Stakeholders require the ability to participate effectively and to accept various roles and responsibilities. Such participation presupposes certain skills and abilities (social, cultural, financial). Landscape processes require competent and effective representation, and institutions that are able to engage with all the issues raised by the process.

Note: These principles have been adopted by the Subsidiary Body on Scientific, Technical and Technological Advice of the Convention on Biological Diversity, and were produced by an intergovernmental and interinstitutional process.

Source: adapted from Sayer et al. (2013).

and broad stakeholder and institutional support. These factors must therefore be carefully integrated into all stages of planning and implementation. Even with the best of intentions, however, following through with effective planning and implementation of restoration requires good governance, and an equitable distribution of the costs and benefits of restoration interventions (van Oosten 2013).

1.5 Scope and organization of this report

This document provides a comprehensive review of support tools to guide decision making prior to and during the restoration process.

In Chapter 2, we describe tools for taking the first steps in the restoration process (Figure 2).


Typically, these tools would be used in the first stage, after commitments to undertake restoration have been made. This stage involves assessing the key success factors that are in place, and assessing opportunities for restoration (IUCN and WRI 2014; WRI 2015). This process also helps to identify key stakeholders and key knowledge sources who should be engaged in the process. An example of a tool used at this stage is the Restoration Opportunities Assessment Methodology (ROAM; IUCN and WRI 2014), which also guides the selection of target areas for restoration, where more detailed spatial information is gathered and assessed. ROAM involves defining and mapping opportunities for restoration within targeted areas. Restoration planning can also utilize data on local governance, community initiatives and other social considerations to determine whether restoration interventions are even necessary, or what objectives are most appropriate and desirable.

Chapter 3 focuses on decision support tools developed to assess the supply of ecosystem services and potential for biodiversity conservation through restoration interventions. These include tools for assessing and mapping ecosystem service supply and delivery to stakeholders and for modeling scenarios of their production following proposed

restoration interventions. These tools can also be used to analyze synergies and trade-offs between different ecosystem services, providing essential guidance for spatial prioritization, and for matching restoration interventions to achieve specific outcomes that benefit particular stakeholder groups. This chapter also reviews decision support tools for incorporating biodiversity values in systematic restoration planning, and for assessing and mapping restoration costs and cost-effectiveness.

In Chapter 4, we describe decision support tools for spatial planning of specific restoration interventions based on biophysical conditions, socioeconomic conditions, and stakeholder needs and preferences, as defined by the landscape approach and the principles of FLR. These approaches are based on spatial prioritization tools that incorporate multiple biophysical and social criteria. Tools for selecting tree species that are appropriate to the regional or local conditions are also described.

Chapter 5 concludes our review with an evaluation of the challenges and gaps in use of decision support tools for FLR. We provide a future outlook and recommendations for the enhancement of decision support tools and expansion of their access and use in planning and implementing FLR throughout the world.

Figure 2. Roadmap of this review.

Chapter 2Tools for preparation and assessment• De�ning objectives• Taking stock and gathering data• Assessing restoration opportunities and priorities• Assessing opportunities for �nancing• Engaging stakeholders• Building capacity

Chapter 3Tools to evaluate potential restoration outcomes• Assessing supply, spatial distribution and economic value of ecosystem services• Assessing distribution of bene�ts among stakeholders• Assessing bene�ts of restoration for biodiversity• Assessing synergies and trade-o�s among restoration outcomes• Estimating restoration costs and cost-e�ectiveness• Building capacity

Chapter 5Conclusions: Synthesis and outlook• Visualising change• Theory of change and adaptive management• Frontiers for decision support tools• Operationalizing FLR

Chapter 4Tools for prioritization, spatial planning and species selection• Systematic restoration planning• Multi-criteria spatial prioritization• Selecting restoration interventions• Selecting appropriate species

In 2016, the small (118,484 km2) landlocked African country of Malawi, made a commitment to the Bonn Challenge (http://www.bonnchallenge.org/content/challenge) to restore 2 million ha of forest by 2020, and an additional 2.5 million ha by 2030. Among the world’s least-developed countries, Malawi’s economy is based on agriculture, and 85% of the population lives in rural areas. Tobacco, sugarcane, cotton, tea, corn, potatoes, sorghum, cattle and goats are the principle agricultural products. Agriculture is largely rain-fed and accounts for more than one-third of the GDP and 90% of export revenues. Miombo woodlands once covered Malawi, but the trees have low commercial value and deforestation is a serious problem. Between 1990 and 2010, the country lost 17% of its Miombo forests due to fuelwood collection, subsistence agriculture, and expansion of commercial agriculture (FAO 2010). Forests on customary land owned by smallholders account for 63% of the forest area. In January 2015, an estimated 64,000 ha of cropland were washed away in southern Malawi due to devastating floods. More than a million people across the country were affected, including 336,000 who were displaced and over 100 people who perished. These conditions created strong motivation to reverse the devastating effects of deforestation and land degradation by developing a plan for FLR.

Malawi’s National Forest Landscape Restoration Assessment was completed in June 2017 (Malawi Ministry of Natural Resources 2017). The country’s National Forest Landscape Restoration Strategy focuses on five types of interventions: conservation agriculture, intensive agroforestry, farmer-managed natural regeneration, community plantations and private woodlots, and natural forest management and watershed protection, covering a total of 4.5 million ha (Republic of Malawi 2017). It is critical to Malawi’s national interests to plan FLR interventions that enhance rural livelihoods and

agricultural production. How did Malawi begin the process of assessing where and how different forms of restoration can take place to meet their needs and objectives?

Many countries find themselves in a similar situation to Malawi, but each has arrived at the starting gate with a unique deforestation and development history, political system, biogeographic setting and socioeconomic realities. In this chapter, we review tools that are being used to begin the process of FLR within countries or regions. These tools are designed to help restoration actors become prepared, to assess their level of readiness, and to begin planning where and how restoration interventions should be located and financed within the country. Some tools can also be used to develop baseline data and monitoring protocols.

The tools discussed in this chapter are designed to guide the first stages of FLR planning and cooperation at a coarse scale, such as within regions or entire countries. But few of these tools are really designed to plan and implement fine-scale restoration interventions on the ground. Rather, the frameworks and tools presented here provide a foundation for the long-term work that needs to be done to transform landscapes and bring social and environmental benefits to many stakeholder groups for many decades into the future.

2.1 A diagnostic tool for identifying readiness for FLR

Based on an analysis of 16 case studies and peer-reviewed literature, the World Resources Institute (2015) developed a tool called the Restoration Diagnostic for assessing the status of key success factors for FLR. The case studies revealed that successful restoration processes share three properties or themes. First, decision

2 Tools for preparation and assessment


makers, landowners or individual citizens were inspired or motivated to catalyze the FLR process. Second, a set of ecological, market, policy, social and institutional conditions helped create a favorable environment for FLR. Third, resources and capacity were mobilized to implement FLR at a sustained level on the ground (WRI 2015). Each of these themes can be broken down to particular success factors that, when present, create possibility, momentum and integration of FLR within national priorities (Table 3).

The Restoration Diagnostic is a tool (or method) for developing national or regional strategies for successful FLR based on a three-step process:1. Users define the scope, jurisdiction or

geographic region for analysis. This can be an entire country, a department or state, or a watershed or basin.

2. Users conduct an assessment to identify the success factors that are present and those that are missing within the target area.

3. Users identify particular policies, incentives or practices to address the missing factors.

This tool can be used by many different groups or individuals, including managers of non-governmental organizations, government agencies, communities or landowners. Development agencies can use the diagnostic to develop ways to enable FLR under financing schemes. Companies can use the diagnostic as a planning tool for their own voluntary or mandated restoration projects. The diagnostic tool can be adapted for use at different jurisdictional scales. Although the restoration diagnostic is being used in sub-national assessments, it needs further adaptation to become a useful tool for landscape-scale assessments.

The Restoration Diagnostic tool was utilized by WRI and IUCN in Rwanda in 2013 and was designed to be used as a component of national scale restoration assessments. In Step 1, the decision was made by the President to make the country’s entire rural area – 2.1 million hectares – the scope of the diagnostic, matching the area committed to the Bonn Challenge. This area constituted 80% of the country’s total land area. In Step 2, a series of workshops were held with different participant groups to assess the status of key success factors, which highlighted several gaps that needed to be addressed, such as increasing seed availability, improving government coordination and offering extension services (Table 3). In the

third step, participant groups identified strategies to address these gaps, as well as to address success factors that were not fully in place. They then ranked these strategies, prioritizing those with highest urgency and highest feasibility of implementation (Table 3). Among other strategies, the diagnostic process in Rwanda highlighted the need to develop strategies to communicate the benefits of restoration through a public awareness campaign, increase native species in plantings and convene joint sector working groups to coordinate government agencies to prioritize and promote restoration activities (Table 3). The method was also utilized in Brazil’s Atlantic Forest and Ecuador, highlighting particular strategies to improve key success factors for FLR in these unique contexts (WRI 2015).

2.2 A tool for assessing restoration needs and ecosystem conditions

The Forest Landscape Assessment Tool (FLAT) was designed to help local government agencies, nongovernmental organizations, land managers and private landowners in the northwestern United States evaluate forest lands, for purposes of management and stewardship (Ciecko et al. 2016). The tool provides procedures and tools to determine forest ecological conditions and potential threats to enable rapid assessment of baseline conditions, determine and prioritize restoration needs across a landscape, and to monitor progress in reaching land management goals. The assessment is conducted in three phases: (1) forestry type mapping, using aerial imagery and classification of areas into management units; (2) a ground-based field assessment of management units; and (3) development of management strategies for each management unit. The framework of this landscape-scale approach is flexible and can be used by different types of landowners and across a wide range of forest ecosystems. FLAT is currently being used to develop and implement forest stewardship plans within King County in Washington State, USA. Future assessments using FLAT can be used to monitor progress and evaluate effectiveness of restoration interventions. The approach does not incorporate land uses outside of natural forests and does not explicitly incorporate economic or livelihood criteria for prioritization of restoration (Ciecko et al. 2016).


THEME FEATURE KEY SUCCESS FACTOR RESPONSE

MOTIVA

TE

BENEFITS

Restoration generates economic benefits

Restoration generates social benefits

Restoration generates environmental benefits

AWARENESSBenefits of restoration are publicly communicated

Opportunities for restoration are identified

CRISIS EVENTS Crisis events are leveraged

LEGAL REQUIREMENTS

Law requiring restoration exists

Law requiring restoration is broadly understood and enforced

ENABL

E

ECOLOGICAL CONDITIONS

Soil, water, climate and fire conditions are suitable for restoration

Plants and animals that can impede restoration are absent

Native seeds, seedlings or source populations are readily available

MARKET CONDITIONS

Competing demands (e.g. food, fuel) for degraded forestlands are declining

Value chains for products from restored area exist

POLICY CONDITIONS

Land and natural resource tenure are secure

Policies a�ecting restoration are aligned and streamlined

Restrictions on clearing remaining natural forests exist

Forest clearing restrictions are enforced

SOCIAL CONDITIONS

Local people are empowered to make decisions about restoration

Local people are able to benefit from restoration

INSTITUTIONAL CONDITIONS

Roles and responsibilities for restoration are clearly defined

E�ective institutional coordination is in place

IMPL

EMEN

T

LEADERSHIPNational and/or local restoration champions exist

Sustained political commitment exists

KNOWLEDGE

Restoration “know-how” relevant to candidate landscapes exists

Restoration “know-how” transferred via peers or extension services

TECHNICAL DESIGN

Restoration design is technically grounded and climate resilient

Restoration limits 'leakage'

FINANCE AND INCENTIVES

Positive incentives and funds for restoration outweigh negative incentives

Incentives and funds are readily accessible

FEEDBACKE�ective performance monitoring and evaluation system is in place

Early wins are communicated

Launch a public awareness campaign to highlight the benefits of a diverse range of trees, especially native species.

Introduce target to plant at least 20% with native species.

Build Tree Seed Center capacity to supply the quantity, quality, and diversity of seeds, especially native species.

Mandate one agency to promote and provide technical guidance on agroforestry.

Use Joint Sector Working Group to coordinate government agencies and help prioritize/promote restoration activities.

Improve understanding among ministerial and district sta� of how small-scale landownersmanage their woodlots to identify acceptable measures for improving production.

Improve existing district- and sector-level extension services by aligning performance targets of agriculture and forest sta� with restoration goals.

STRATEGIES

In place

Partly in place

Not in place

Note: For each key success factor, responses indicate whether the factor is in place (green), partly in place (yellow) or not in place at all (red). Strategies to overcome the major limitations in key success factors are indicated.

Table 3. Results of the restoration diagnostic in Rwanda.


2.3 A tool for mapping, monitoring and planning at the landscape scale

A guide to spatial planning and monitoring at the landscape scale was designed by EcoAgriculture Partners in collaboration with TerrAfrica, to guide diverse stakeholders in locating, designing and monitoring interventions to improve benefits for people living in rural landscapes (Willemen et al. 2014). This guide serves as a tool for integrating agricultural production, biodiversity conservation and livelihood security. Stakeholders undergo an eight step process, conducted in five phases (Figure 3).

The first phase is devoted to stocktaking, and identifying and locating benefits within the landscape, such as water supply and regulation, habitat provision, crop production, and moderation of extreme climate events. Stakeholders develop a GIS platform composed of global- and national-scale maps, to identify important areas that correspond to the supply of landscape benefits and to inform planning of place-based interventions. Areas are identified where changes would lead to improvement of landscape benefit flows. A monitoring element

is included to guide stakeholders to quantify and describe landscape benefits, so that change can be detected and measured.

Desired landscapes changes are then negotiated, planned, and implemented in phases 2–4. Phase 5 involves development of a spatial strategy for monitoring and evaluation of landscape changes that also incorporates adaptive management of landscape interventions. This approach was applied in a case study in collaboration with TerrAfrica in the lower Burqua Abagabir Wejig micro-watershed in southern Tigray, Ethiopia, to support future SLM activities in this region (Willemen et al. 2014). SLM/ILM objectives, activities, and underlying principles greatly overlap with those of FLR (as covered in Chapter 1).

2.4 A tool for assessing FLR opportunities and approaches at national or sub-national levels

Global-scale assessments of opportunities for restoration are not adequate for making national or sub-national opportunity assessments, nor

Plan landscape changesPhase 3

Assess how change in the landscape impacts people

their locations in the landscape

Identify areas where change for improved

Identify who currently manages these areas

measure changes in

Understand the landscape

Design landscape changes at selected locations

Implement landscapes changes at selected locations

Monitor and evaluate changes in the landscape

1

2

3

4

5

6

7

8

Phase 1

Negotiate desired landscape outcomes

Phase 2

Implement landscape changes

Phase 4

Evaluate the landscapePhase 5

Figure 3. The eight steps of planning and monitoring landscape interventions, in which maps are used within the five-phase adaptive collaborative management process.

Source: Willemen et al. (2014).


for defining target areas for specific types of restoration activities. To define and implement national commitments to the Bonn Challenge, the United Nations Convention on Biological Diversity (UNCBD), United Nations Convention to Combat Desertification (UNCCD) and the United Nations Framework to Combat Climate Change (UNFCCC), countries must be able to identify and map the areas to be restored and to project how restoration actions will provide a wide range of environmental and socioeconomic benefits. Ultimately, the goal of these activities is to halt degradation and deforestation processes, and restore functionality to landscapes through a variety of interventions that increase tree cover, restore forest ecosystems and their services, protect biodiversity, reduce poverty and improve sustainability of land use.

The need for making these assessments at national and sub-national levels led to the development of the Restoration Opportunities Assessment Methodology (ROAM) by WRI and IUCN. This methodology grew out of initial experiences working with multiple stakeholders in Ghana in 2011, and a ‘road-test’ edition was published in 2014, informed by further experiences conducting assessments in Rwanda and Mexico (IUCN and WRI 2014). The assessment methodology was designed to evolve over time, as new experiences, contexts and technological developments reveal key aspects that were underdeveloped or required further refinement.

The purpose of ROAM is to lead national or sub-national groups of stakeholders through a guided process to become engaged in stock-taking, mapping, analysis of costs and benefits of restoration activities, and assessment of financing and investment options. The ROAM process is a multi-stakeholder assessment process that provides an opportunity for collaboration across different agencies, sectors and institutions that may not have worked together previously.

To date, opportunity assessments, maps and economic analyses have been completed or are in progress in 24 countries or states around the world (Box 2). It is important to note that ROAM was not developed as a tool for planning or implementation of FLR at landscape scales, although sub-national assessments are being conducted or have been completed in Brazil, Nicaragua, Indonesia, Ethiopia and Mexico.

ROAM is often used in combination with the Restoration Diagnostic (WRI 2015), as both approaches provide key insights into information, technical, and other gaps that must be addressed to plan (and ultimately implement) successful FLR at national and sub-national scales. For example, initial assessment of restoration opportunities in Ghana was hindered by the lack of a recent land-cover/land-use map and recent Landsat mosaic coverage. Technical infrastructure for acquiring and processing data are often major constraints. When spatial data are lacking, a technique called ‘knowledge mapping’ can be used to document local expert knowledge on the most appropriate FLR options to apply to different target areas (Pistorius et al. 2017). The ROAM process, by its very nature, can stimulate efforts to build technical capacity, and to obtain and analyze data for carrying out assessments.

The ROAM process is flexible and can be adapted to address specific needs and constraints within each country or sub-national unit where it is being used. Three main phases are involved: (1) preparation and planning; (2) data collection and analysis; and (3) results to recommendations (Figure 4).

Conducting an assessment typically takes an assessment team about 15–30 workdays, which can be spread out over a 2–4-month period. All phases require engagement of multiple stakeholders who work together in a workshop setting as a team. The ROAM process can be effectively conducted within sub-national regions, as planning and implementation of FLR interventions are best suited within a more restricted local context, with participation of local stakeholders who know most about the region and who are invested in the success of FLR outcomes. Pistorius et al. (2017) describe the ROAM process that was conducted within the Amhara National Regional State in Ethiopia, a region of 16 million ha, where some of the most severely degraded landscapes within the country are found.

During Phase 1 of the ROAM process, stakeholders prioritize the types of restoration interventions best suited to target areas identified as opportunities. For example, the ROAM handbook describes seven types of FLR interventions in three main categories: (1) forest land (natural or planted); (2) agricultural land (agroforestry and fallows); and (3) protective land and buffer zones (coastal zones,


Box 2. Freely available support tools discussed in this chapter

Restoration Diagnostichttps://www.wri.org/sites/default/files/WRI_Restoration_Diagnostic_0.pdfhttps://www.youtube.com/watch?v=V-s2NNcOCLA

Forest Landscape Assessment Tool (FLAT): Rapid Assessment for Land Managementhttps://www.fs.fed.us/pnw/pubs/pnw_gtr941.pdf

Spatial Planning and Monitoring of Landscape Interventions: Maps to Link People with their Landscapes. A User’s Guidehttps://ecoagriculture.org/publication/a-landscape-perspective-on-monitoring-evaluation-for-sustainable-land-management/spatial-planning-and-monitoring-of-landscape-interventions-maps-to-link-people-with-their-landscapes/

Online Step-by-step ROAM Guide http://www.forestlandscaperestoration.org/tool/roam-online-step-step-guide

ROAM Handbook in Englishhttp://www.forestlandscaperestoration.org/tool/roam-online-step-step-guide

ROAM Handbook in Frenchhttps://portals.iucn.org/library/node/45771

ROAM Handbook in Spanishhttps://portals.iucn.org/library/node/45770

ROAM Handbook in Portuguesehttps://portals.iucn.org/library/node/45772

ROAM Handbook in Russianhttps://portals.iucn.org/library/node/45872

ROAM Handbook in Indonesianhttps://portals.iucn.org/library/node/46351

FAO Sustainable Financing for Forest and Landscape Restorationhttp://www.fao.org/3/a-i5031e.pdf

IUFRO Forest Landscape Restoration as a Key Component of Climate Change Mitigation and Adaptationhttp://www.iufro.org/publications/series/world-series/article/2015/12/01/world-series-vol-34-forest-landscape-restoration-as-a-key-component-of-climate-change-mitigation/

Attracting Private Investment to Landscape Restoration: A Roadmaphttp://www.wri.org/publication/attracting-private-investment-to-landscape-restoration

Implementing Forest Landscape Restoration, a Practitioner’s Guide. Stanturf J, Mansourian S and Kleine M, eds. 2017. Vienna, Austria: International Union of Forest Research Organizations, Special Programme for Development of Capacities (IUFRO-SPDC). 128.https://www.iufro.org/science/special/spdc/flr/flr/pract-guide/

Free Prior and Informed Consent: An Indigenous Peoples’ Right and a Good Practice for Local Communities. Manual for Project Practitioners. FAO. 2016. 52.http://www.fao.org/documents/card/en/c/5202ca4e-e27e-4afa-84e2-b08f8181e8c9/

watershed protection or erosion control). Different criteria and indicators can be used to select and map restoration opportunities, including status of

biophysical properties of soil, levels of disturbance and fragmentation of forests and watersheds, availability of land, land ownership, competing

https://www.wri.org/sites/default/files/WRI_Restoration_Diagnostic_0.pdf

https://www.youtube.com/watch?v=V-s2NNcOCLA

https://www.fs.fed.us/pnw/pubs/pnw_gtr941.pdf

https://ecoagriculture.org/publication/a-landscape-perspective-on-monitoring-evaluation-for-sustainable-land-management/spatial-planning-and-monitoring-of-landscape-interventions-maps-to-link-people-with-their-landscapes/



http://www.forestlandscaperestoration.org/tool/roam-online-step-step-guide

http://www.forestlandscaperestoration.org/tool/roam-online-step-step-guide

https://portals.iucn.org/library/node/45771





http://www.fao.org/3/a-i5031e.pdf

http://www.iufro.org/publications/series/world-series/article/2015/12/01/world-series-vol-34-forest-landscape-restoration-as-a-key-component-of-climate-change-mitigation/

http://www.iufro.org/publications/series/world-series/article/2015/12/01/world-series-vol-34-forest-landscape-restoration-as-a-key-component-of-climate-change-mitigation/

http://www.wri.org/publication/attracting-private-investment-to-landscape-restoration

https://www.iufro.org/science/special/spdc/flr/flr/pract-guide/

http://www.fao.org/documents/card/en/c/5202ca4e-e27e-4afa-84e2-b08f8181e8c9/


economic interests, livelihood improvement, costs of interventions, benefits for ecosystem service flows, and alignment with existing policies and laws. Once criteria are selected and opportunities have been identified, particular restoration interventions can be assigned to them according to their suitability for addressing a range of social and environmental needs. Multiple stakeholders, including experts, should be involved in selecting the criteria for prioritizing restoration areas (Orsi et al. 2011). Orsi et al. (2011) describe a process for identifying criteria and indicators based on an open, unstructured survey followed by a face-to-face meeting. These indicators include factors related to biodiversity conservation as well as feasibility, but exclude socioeconomic factors.

The second phase of ROAM involves core activities involving the collection and analysis of different forms of data (Figure 4). Available technical expertise and data vary widely across different countries, so types and depth of analyses will also vary widely. The goal is to use the best information available to identify areas with the highest restoration potential, to determine which restoration interventions are most appropriate

for these areas and to assess the costs and benefits of these activities. The analysis process should consider the needs of key end-users and the need to involve expertise and perspectives from different sectors, including agriculture, forestry, water, energy, gender, economics and development. As countries gain data and technical expertise, these analyses can be revisited to improve the evidence base for restoration planning.

Restoration of degraded land offers the greatest potential for economic and environmental benefits, but the cost of restoration is high, as degraded lands offer limited potential for low-cost forest restoration through natural regeneration or fallow management. The greater the extent of ecological disturbance, the greater is the restoration debt (Ghazoul and Chazdon 2017; Moreno-Mateos et al. 2017). The causes of land degradation must also be addressed to avoid wasted efforts and reduce the future need for restoration interventions. Likewise, care must be taken in assessing degraded forest areas to avoid classifying native grasslands or wetland ecosystems as degraded forest (Veldman et al. 2015). Ecologists or land managers with knowledge of non-forest ecosystems should

Figure 4. The phases of the Restoration Opportunity Assessment Methodology.

Source: Pistorius et al. (2017).


be consulted in making these assessments, particularly if they are based heavily on remote sensing data. In regions where natural tree cover is sparse and intermixed with grassland formations, clarity is often lacking regarding how to define land covers such as ‘open forests’ and ‘woodlands’, and the role of natural disturbance regimes that maintain the diversity of these open structure within landscapes (Veldman et al. 2015; Holmgren and Scheffer 2017). Key questions can be useful in making decisions regarding the optimal levels and types of forest cover in landscapes in the context of FLR (Table 4).

Assessing forest degradation or disturbance is not as straightforward as assessing deforestation where forests have been replaced by non-forest land use, or land degradation where agricultural productivity is minimal or zero (Ghazoul and Chazdon 2017). Perceptions of forest degradation are social constructs, subject to cultural and ethical norms, and conflicts among stakeholders can arise when forest areas are labeled as degraded. Depending on the historical and ecological benchmarks used, for example, drivers of deforestation and forest degradation, such as increased fire frequency, might be considered beneficial for the rejuvenation and restoration of native grassland ecosystems. For these reasons, local knowledge and maps, with detailed forest and agricultural land-use history, provide critical information for identifying areas that are priorities for specific forest restoration interventions. Likewise, knowledge of local, state, and national policies and incentives is essential in spatial planning, and engaging farmers, landowners and communities in restoration activities on private or customary land.

Selecting the types of restoration interventions to apply in particular areas requires broad stakeholder engagement and matching of restoration objectives with appropriate social and environmental conditions on the ground. For example, in the Rwanda ROAM assessment, five broad interventions were identified and opportunity areas for each type were mapped (Rwanda Ministry of Natural Resources 2014). Agroforestry interventions compose 45% of the total restoration opportunity, and consist of two types of interventions: on steeply sloping land; and on flat or gently sloping land. A third intervention is improved woodlot and timber plantation management, which is suitable for

11% of the total restoration opportunity in Rwanda (256,000 ha). These plantations will be used for fuel wood and timber production. Protection and restoration of natural forests is the fourth intervention, suitable for 1% of the total restoration opportunity. The fifth intervention is protective forests for preventing erosion on steep slopes and hillsides and to plant buffer zones along rivers and wetlands, which covers 5.1% (122,540 ha) of the total restoration opportunity in Rwanda (Rwanda Ministry of Natural Resources 2014).

Given the planned allocations of different restoration interventions, an economic and financial analysis is then undertaken to assess the costs and benefits of restoration transitions, and to estimate return-on-investment (ROI) at a national or regional scale (Ding et al. 2017). The economic analysis requires the application of available information on the costs of implementation, land opportunity costs and the benefits of restored land, such as timber yields, crop yields, prevention of erosion and carbon sequestration. In the Rwandan example, the costliest transition is from traditional agriculture to agroforestry, followed by planting of protective forests. Managing woodlots and natural regeneration are the least costly restoration options. A carbon abatement curve is used to illustrate the potential for carbon sequestration through restoration interventions (Figure 5). This curve was made by estimating the the total CO2 sequestered over 20 years for five different restoration transitions and calculating the net present value per ton of CO2 for each transition.

Table 4. Four key questions to provide information needed to decide on the optimal tree cover of a landscape.

• How does tree cover and composition affect the status of supporting, provisioning, regulating and cultural ecosystem services?

• How does the resilience of ecosystem services under climate change vary with tree cover?

• How do socioeconomic and cultural backgrounds mediate the perception of ecosystem services of tree cover?

• What are the ecological and social bridges and barriers to manage changes in tree cover pro-actively in the face of climate change?

Source: Holmgren and Scheffer (2017, Table 1).


The greatest potential for carbon storage is provided by the transition from traditional agriculture to agroforestry, which would store an additional 31 Mt of CO2 over 20 years. Further, each ton of carbon stored through agroforestry would bring financial profits of RWf 17,000 from crop yields, reduced erosion and wood production (Figure 5). Whether or not these estimates are accurate, they provide a relative assessment of the costs and benefits of different interventions, and are essential sources of information for planning large-scale activities at a national or sub-national scale.

A map of restoration opportunities is another product of a ROAM assessment that provides critical information for planning and decision-making at national and sub-national scales. The restoration opportunities map produced for the Guatemala ROAM assessment shows areas selected for agroforestry with annual crops, silvopastoral systems, agroforestry with permanent crops, production forests, conservation forests, and riparian forests (Figure 6). The restoration opportunities map provides a vision of the future, and reflects a consensus among multiple stakeholders of the relevant criteria deemed to be important in this selection. It is important to revisit these maps and reassess restoration priorities periodically, as more information becomes available and as local and national priorities and opportunities change.

Figure 5. A carbon accrual curve for Rwanda based on their ROAM assessment.

Source: Figure 14 from Rwanda ROAM assessment.

Optimization modeling can also be a tool for prioritization of restoration that incorporates both spatial and temporal dynamics (Chapter 4). A dynamic prioritization model can provide a schedule for where and when restoration should take place and provides operational guidance and support for cost-effective restoration planning (McBride et al. 2010; Wilson et al. 2011).

Economic analyses tools can also guide decision making for FLR. Ding et al. (2017) outline three key steps in conducting economic assessments based on specific types of restoration interventions. The first step is to model changes in ecosystem service supply resulting from restoration interventions (see Chapter 3). These models should compare restoration scenarios with status quo (unrestored) scenarios to highlight the changes in ecosystem service supply and the economic benefits emerging from restoration interventions. Next, the costs and benefits of restoration interventions are estimated for these scenarios using market and non-market valuation analyses. Costs include direct costs of restoration as well as the opportunity costs of foregoing agricultural use of the land. The third step is to conduct economic analysis such as cost–benefit analysis, cost-effectiveness analysis, spatial restoration optimization analysis and macroeconomic analysis. The results of these analyses will help to prioritize investments in different restoration projects and to determine


policies for incentives or subsidies (Ding et al. 2017). Currently, however, no standardized tools exist to apply economic valuation methods and to help restoration practitioners develop economic scenarios for particular regions or landscapes to guide decision making and predict financing needs.

2.5 Tools for seeking financing for FLR

During the third phase of ROAM, the outputs of the assessment team are validated and discussed with key decision makers and other stakeholders to generate the needed financial and political support.

Figure 6. Map of restoration opportunities and associated planned interventions from the Guatemala ROAM assessment.

Source: Figure 15 from ROAM Handbook.

Protected areas Non selected areasAgroforestry with annual cropsSilvopastoral systemsAgroforestry with permanent crops

Forests for productionForests for conservationRestoration in protected areasRiparian forests

MangrovesWetlands


At this point, the ROAM process has led to a series of recommendations and detailed plans, but actual implementation requires financing, investment plans and packaging incentives to landowners or communities. These tools are not adequately addressed by ROAM and additional support tools for financing FLR are needed.

Financing for FLR can come from public or private sources, or a combination of these (FAO and UNCCD 2015; Ding et al. 2017; Faruqi and Landsberg 2017). Faruqi and Landsberg (2017) provide a roadmap tool for practitioners seeking to attract private investment for FLR. The roadmap targets restoration stakeholders who have limited experience with investment, and who are looking to attract between USD 500,000 and USD 10 million in private capital. The roadmap highlights four steps in the process: (1) identify and engage private investors that are best suited to particular restoration projects; (2) develop a compelling and robust business model; (3) set up an investable entity to channel funding; and (4) track financial and operating performance (Faruqi and Landsberg 2017). The roadmap document also provides guidance on what goes into a restoration business model and how to track operating performance.

Simula (2008) outlined three investment steps for sustainable financing: (1) initial up-front financing to cover transaction costs and FLR project design; (2) financing and investments for operational costs of implementation of restoration interventions; and (3) sustained financing leading to self-sustaining financing of long-term project costs. FAO and UNCCD (2015) provide a list of funding sources and financing approaches for FLR. Financing for FLR can also utilize compliance or voluntary markets for carbon mitigation, following the model of REDD+ financing (FAO and UNCCD 2015; Stanturf et al. 2015). Additional approaches for funding for FLR include crowd funding, green bank cards and restoration bonds (FAO and UNCCD 2015).

2.6 Tools for engaging stakeholders in the FLR process

One of the basic principles of FLR is to respect the rights of all peoples involved (Box 1). The principles of Free Prior and Informed Consent (FPIC) apply to indigenous peoples in many countries where land ownership and tenure

rights are often vaguely defined. The FAO Free, Prior and Informed Consent (FPIC) Manual is a tool for project managers and practitioners that provides information about the right to FPIC and how it can be implemented in six steps (FAO 2016). Planning FLR requires recognition of local knowledge and information systems through active stakeholder participation. As there may well be competing demands on land and differing views regarding what types of restoration interventions to undertake, it is essential to reconcile these views as much as possible. Several tools have been developed to help in this process (Lynam et al. 2007). These tools have mostly been applied in decision making regarding local participation in forest use and conservation.

Despite the importance of identifying actors and power relationships among stakeholders, few tools are available for mapping and understanding the dynamics of social landscapes compared to biophysical landscapes. Stakeholder mapping tools are useful for identifying stakeholders, defining their stakes and potential impacts in the restoration effort, and engaging them in dialogue (Stanturf et al. 2017). Several tools are available that could be applied to FLR contexts, including the Alignment, Interest, and Influence Matrix developed by the Overseas Development Institute (Mendizabal 2010) and the Participatory Impact Pathway Analysis, which uses network mapping to explore how stakeholders are linked to and influence each other (Douthwaite et al. 2007).

A recently developed guidebook (Buckingham et al. 2018) focuses on the networks of actors within the landscapes in the context of FLR. The guidebook centers on two main approaches: (1) mapping actors’ resource flows and (2) mapping actors’ priorities and values. This social mapping methodology has been tested in India, Indonesia, Kenya, Mexico, Rwanda and Brazil. By using this guidebook, restoration practitioners can be more efficient with resources, collaboration and outreach, and better anticipate potential conflicts and bottlenecks in planning and implementing restoration interventions.

Engaging stakeholders and sustaining their active participation in all phases of the FLR process can be challenging and time-consuming. Stakeholders can be classified into three different groups: (1) primary (direct) stakeholders, who live or work in the landscape or in downstream areas within the


watershed; (2) secondary (indirect) stakeholders, such as government agencies and decision makers, who are responsible for governing and managing resource- and land-use; and (3) interest groups, such as national experts, NGOs or international organizations, who have a stake in restoration activities and outcomes. As stakeholder become engaged in FLR decision-making, the governance of the landscape changes. There is no unambiguous framework for how to best guide the process of decision making and implementation of FLR. Analysis of case studies from Indonesia suggests that a flexible approach to social learning is likely to be more successful than a strongly institutionalized approach based on rigid design criteria (van Oosten et al. 2014). Recent experiences in Brazil further recommend incorporation of social science methods during project design (Ball et al. 2014), an aspect that tends to be overlooked. A recent country-wide assessment in Colombia revealed a notable lack of local participation during design and implementation of forest restoration projects (Murcia et al. 2016). Better access and training for using stakeholder engagement tools can potentially remedy this situation at both national and landscape scales.

Encouraging people to visualize future landscape scenarios through drawings or computer simulations can be a highly effective approach to engaging their participation in restoration decisions and activities (Chapter 5; Boedhihartono 2012). Role-playing games and participatory scenario planning illustrate emerging approaches for stakeholder engagement, but are yet to be applied to FLR contexts (Garcia Barrios et al. 2011; Oteros-Rozas et al. 2015; Garcia et al. 2016). FLR presents a promising new frontier for developing scenarios

and role-playing games that help to reconcile trade-offs and conflicts among stakeholders, and engage them in envisioning alternative futures (Metzger et al. 2017).

2.7 Concluding remarks

Undertaking regional and national assessments of restoration and financing opportunities creates a favorable environment for multi-sectorial planning and engagement of key stakeholders – essential conditions for the beginning the implementation of successful restoration actions at local and national scales, and for the mobilization of sustained financial and public support. Recognizing and creating an enabling environment for FLR are essential first steps. Enabling conditions for FLR need to be manifest at different scales of geography and governance. A combination of tools are needed to catalyze FLR that help to build a unified vision of restoration opportunities and appropriate interventions, and that involve all relevant stakeholder groups in the process from the earliest stages. The tools discussed in this chapter can be useful for supporting decision making during the preparation and assessment stages of FLR, but they are generally focused on national scales and on coarse-level assessments. Tools focused on catalyzing and implementing FLR at landscape scales have yet to be fully developed and applied in the tens or hundreds of thousands of landscapes in need of restoration. To be effective in enabling FLR implementation on the ground, these tools needs to be accessible and easy to use by FLR practitioners and stakeholders and require relatively little training and supervision.

Underlying the multiple objectives of FLR is the need to increase the provision of ecosystem services in ways that benefit local livelihoods and promote the conservation of biodiversity. In addition, these objectives need to be met while reducing the overall costs of planning and implementation of restoration interventions. This chapter focuses on decision support tools developed to inform how restoration interventions will provide the greatest impact on ecosystem services and biodiversity conservation, at minimal cost. These layers of information are derived from spatially explicit local or global data on land cover, topography, climate, species distributions and habitat requirements, household surveys, or other social or biophysical datasets. Once assembled, these layers can be used as inputs for multi-criteria prioritization analyses (Chapter 4), analysis of alternative scenarios of restoration interventions, and to evaluate trade-offs and synergies among different ecosystem services, biodiversity and costs. Individual layers can also be used to address specific restoration objectives. Valuation of ecosystem services is critical for assessing market values, returns on investment and for planning payments for ecosystem service programs.

Models predict outcomes based on input data, but they vary in the way they process data, the type of data used and the nature of the outputs. In applying modeling tools (software packages with models as components) to decision making for systematic restoration planning, several considerations are important. The selection of tools and models for mapping data layers depends heavily on the main objectives of restoration interventions, the types of data available, and the skills and capacity to run and apply the selected models (Christin et al. 2016; Bullock and Ding 2018). Which is the most appropriate tool to use, or should more than one tool be used? Are coarse-scale patterns sufficient for decision making? How complex and detailed does the model need to be? Will decisions be made solely

on biophysical data, or will surveys of social and cultural preferences, livelihood needs or economic status be included as well? Bullock and Ding (2018) provide a useful guide with detailed steps to help select the most appropriate modeling tool and models provided by these tools for use in a particular decision-making context. Local stakeholders can become engaged in different stages of the modeling process, such as providing knowledge and developing and testing scenarios. Participatory modeling can build stakeholder capacity, creating consensus and trust, and increasing the likelihood of local engagement in planning and implementation of FLR (Turner et al. 2016).

3.1 Tools and models for assessing and mapping ecosystem service supply

Ecosystem services are the processes and conditions derived from ecosystems that sustain and enhance human wellbeing (Daily 1997; MEA 2005; Reyers et al. 2013). The Millennium Ecosystem Assessment (MEA 2005) brought the concept of ecosystem services to the central stage of the global environmental policy arena. The MEA classification comprised four categories of ecosystem services: supporting, regulating, provisioning and cultural. More recently, the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES 2017) presented a new conceptual framework linking nature and human well-being in Nature’s Contributions to People (NCP) (Pascual et al. 2017a). This new framework incorporates views from social science and humanities, and consists of 18 NCPs in three major categories: regulating, material and non-material (IPBES 2017). These NCPs provide not only benefits or positive contributions, but can also produce negative contributions or losses that affect people (Pascual et al. 2017a). Cultural ecosystem services are implicitly embedded in the NCPs, rather than included as a separate category.

3 Tools to evaluate potential restoration outcomes


Regardless of the framework used, it is important to distinguish among four related components: (1) the supply or stock of ecological properties or functions based on biophysical processes; (2) the actual service derived from these properties or functions that is used by people; (3) the market or non-market value of the ecosystem service; and (4) the change in human well-being that ecosystem services bring (Tallis and Polasky 2009; Tallis et al. 2012). Distinguishing between components one and two is extremely important; Bagstad et al. (2014) found that, with the exception of carbon sequestration, regions that actually provided ecosystem services that were directly connected to beneficiaries amounted to 16–66% of those theoretically capable of supplying ecosystem services. Delivery of the benefits of ecosystem services to people requires the capacities of

individuals (human capital), but also their societies (social capital) and their constructed environment (built capital) (Turner et al. 2016). In most cases, ecosystem service delivery requires co-production by local users (Pascual et al. 2017b). For example, irrigation ditches are dug by human labor to deliver the benefits of water for agricultural production.

Decision support tools to assess the supply of ecosystem services became available during the first decade after publication of the MEA and have since been actively proliferating and evolving (Bagstad et al. 2013; Martinez-Harms et al. 2015). Here, we focus on a selection of decision support tools that show particular use in systematic restoration planning (Chapter 4) and that are currently being used to assess, value and map ecosystem services of highest priority for countries and regions (Table 5).

Table 5. Decision support tools for assessment and mapping of ecosystem services.

Name of tool Description Source Reference

Social Values for Ecosystem Services (SolVES)

SolVES assesses, maps and quantifies the perceived social (non-market) values of ecosystem services.

https://solves.cr.usgs.gov Sherrouse et al. (2014)

Artificial Intelligence for Ecosystem Services (ARIES)

An integrated ecosystem services modeling methodology that gives equal emphasis to service production, flow and use by society, and that acknowledges dynamic complexity and uncertainty.

http://aries.integratedmodelling.org/

Villa et al. (2014)

Integrated Valuation of Ecosystem Services and Trade-offs (InVEST)

InVEST is a set of models spanning terrestrial, freshwater, and marine environments, that use production functions to estimate changes in biodiversity and ecosystem services under different demographic, land-use and climate scenarios.

https://www.naturalcapitalproject.org/invest/

Tallis and Polasky (2009)

Toolkit for Ecosystem Service Site-based Assessment (TESSA)

A practical suite of tools for measuring and monitoring ecosystem services at a site scale and for evaluating the magnitude of benefits that people obtain from them currently, compared with those expected under alternative land-uses.

http://www.birdlife.org/datazone/info/estoolkit

Peh et al. (2013)

Multiscale Integrated Models of Ecosystem Services (MIMES)

A multi-scale, integrated set of models that assess the value of ecosystem services and allow decision-makers to quickly understand dynamics of ecosystem services, how ecosystem services are linked to human welfare and how the value might change under various management scenarios.

http://www.afordablefutures.com/home

Boumans et al. (2015)

Co$ting Nature A web-based tool that jointly maps bundled ecosystem service generation and conservation priorities.

http://www.policysupport.org/costingnature

Mulligan et al. (2010)

Restoration Opportunities Optimization Tool (ROOT)

ROOT uses information about potential impact of restoration interventions together with spatial prioritization maps to identify key areas for ecosystem service provision.

https://www.naturalcapitalproject.org/root/

https://solves.cr.usgs.gov

















Based on the restoration commitments eight countries made to the Bonn Challenge, the most desired ecosystem service benefits are carbon storage, biodiversity protection, timber production, water flow, water quality, pollination, pest control, erosion control, food production, fuel wood production, control of invasive species and disaster risk reduction (IUCN 2015). On the socioeconomic side, desirable benefits of restoration are to improve human well-being, alleviate poverty, provide livelihoods and promote economic development (Christin et al. 2016). The benefits of ecosystem services vary widely across different types of planted and natural forests (Baral et al. 2016). When many different types and arrangements of tree cover are found within a landscape, a wide range of ecosystem services can be provided (Figure 7).

Ecosystem service assessments predominantly focus on the quantification and mapping of specific ecosystem services, with less attention paid to evaluating trade-offs, management planning, or evaluation of alternative policy or land-use options (Martinez-Harms et al. 2015). Additionally, temporal and spatial variability in biophysical and social processes that accompany changes in land use and, hence, both the supply and demand of ecosystem services, are often disregarded. Another major limitation is that modeling the potential supply of ecosystem services at a given location

does not provide information on the temporal trajectory required to reach this potential, which can be critically important for restoration planning. This is because the recovery of ecosystem services during forest restoration can take several decades or longer, so model outputs of potential value or supply can provide a misleading prediction of what will actually be available. Past changes in ecosystem service supply, as well as trade-offs and synergies among different types of services, can be assessed by applying models to input data from different time periods (Han et al. 2017).

Decision making for managing ecosystem services often excludes stakeholder consultation and fails to account for social values and preferences (Martinez-Harms et al. 2015). One tool has been developed to incorporate quantified, spatially explicit measures of social values (non-market values) into assessments of ecosystem services. The Social Values for Ecosystem Services (SolVES) tool (Table 5) calculates and maps a value index on a 10-point scale, based on preference surveys of different subgroups (Sherrouse et al. 2011, 2014).

A common approach to predict the supply of ecosystem services uses production functions and land-cover data. Production functions are mechanistic models that predict how multiple ecological variables influence ecosystem service

Figure 7. The types and rates of delivery of ecosystem services provided by natural, semi-natural, planted forest and planted trees outside the forests.

Note: The thickness of the arrows indicates the relative rate of delivery of ecosystem services.

Source: Baral et al. (2016, Figure 1).


supply. An accurate assessment of the provision of an ecosystem service, however, requires that the production functions provided by ecosystems be matched by the demand of beneficiaries. For example, a patch of native vegetation in an agricultural landscape may support healthy bee populations, but these bees are not providing a pollination service to farmers if there are no active agricultural fields requiring bee pollination within their foraging distance (Tallis and Polasky 2009). Combining ecosystem service production functions along with benefit transfer for economic valuation (Costanza et al. 1997) and ecosystem service demand functions can create a powerful tool for generating estimates of the economic value of ecosystem services (Tallis and Polasky 2009). The tool InVEST was developed to enable modeling production functions for a suite of ecosystem services, permitting evaluation of effects of different management approaches, and analysis of trade-offs and synergies among different ecosystem services (Tallis and Polasky 2009). InVEST models use production functions to estimate changes in biodiversity and ecosystem services under different demographic, land-use and climate scenarios (Table 5).

Spatial scale and location are also important factors in selecting models to predict and map ecosystem service supply. When assessing the supply of ecosystem services at large spatial scales and coarse spatial resolution, simpler models perform well, are easier and less costly to use, and provide results that are most useful to decision makers (Ruckelshaus et al. 2015). On the other hand, if accurate assessments of ecosystem service supply, demand and economic values are needed for decision making within a smaller region or watershed, more complex models can provide better results. The spatial scale of analysis and biophysical factors can strongly impact the accuracy of ecosystem service estimates.

Bagstad et al. (2013) identified 17 tools that assess, quantify, model, value and/or map ecosystem services, and compared them using eight evaluative criteria. The two best-known tools in the public domain that can be applied to a wide range of contexts are InVEST and ARIES (Table 5). InVEST uses only deterministic production models for predicting ecosystem service supply. ARIES uses both deterministic and probabilistic models, and quantifies uncertainties in ecosystem service flows. ARIES also uses agent-based models that simulate actions of individuals or groups to quantify ecosystem service flows to specific beneficiaries

(Bagstad et al. 2014; Villa et al. 2014). The probabilistic models generated by ARIES may be more appropriate for use when input data are scarce (Bagstad et al. 2013). The deterministic models of InVEST have been widely evaluated at regional and national scales, and model predictions have been subjected to more rigorous testing (Song et al. 2015; Salata et al. 2017; Redhead et al. 2018).

The tools MIMES, Co$ting Nature and TESSA also provide spatially based, public-domain models for assessing and mapping ecosystem services that can be applied across a wide range of contexts. Although they have been less utilized than InVEST and ARIES, each of these tools offers valuable decision support for restoration planning (Bagstad et al. 2013, Table 5). MIMES uses system dynamics models that, unlike static models based on production functions, account for temporal dynamics and feedbacks among ecological factors, and can output values of ecosystem services over space and time (Boumans et al. 2015). Co$ting Nature is a web-based tool that uses global databases to assess the supply of multiple ecosystem services (in aggregation) for spatial conservation planning. This tool assesses the magnitude and geographic variability of water, carbon, and tourism-related ecosystem services, as used by local and global beneficiaries (Mulligan et al. 2010). TESSA is a toolkit for rapid assessment of the supply of ecosystem services in sites of importance for biodiversity conservation. The tool guides non-specialists through a selection of methods to identify which ecosystem services are most important at a given site, and to evaluate how the supply of these services and their monetary value would change in response to alternative land uses (Peh et al. 2013). Outputs from TESSA can help decision makers evaluate how forest restoration in a particular area would influence the supply of ecosystem services and the distribution of benefits among stakeholders. The aggregated results, however, restrict analysis of synergies and trade-offs among different ecosystem services, as well as links specific beneficiaries to a particular service (Turner et al. 2016).

3.2 Tools for modeling synergies and trade-offs between ecosystem services

Many studies have documented trade-offs between provisioning (e.g. food, fiber) and regulating (e.g. pollination) ecosystem services (Nelson et al. 2009; Sutherland et al. 2016; Woollen et al. 2016) or between restoring forests to reduce risk of


catastrophic fire and timber production (Vogler et al. 2015; Ager et al. 2017). In the USA, Nelson et al. (2008) examined trade-offs between carbon sequestration and species conservation in the Willamette Basin. Their analysis showed that policies aimed at increasing carbon sequestration led to decreased provision of habitats for species conservation and vice versa. Forest restoration scenarios in bottomland hardwood forests of the Mississippi River Basin showed clear spatial trade-offs between configurations targeting wildlife habitat and those targeted at improving water quality (Barnett et al. 2016). A country-level study in Costa Rica showed that biodiversity hot spots have high co-benefits for carbon, water and scenic beauty, but areas with high carbon storage have lower co-benefits for water and scenic beauty (Locatelli et al. 2014). Forest plantations for timber supply more provisioning services per unit area than natural forests, but offer fewer habitats for wildlife and lower levels of biodiversity (Baral et al. 2016). As no single spatial configuration and few areas can deliver high levels of multiple ecosystem services, it is important to consider large areas for landscape-scaling planning, permitting more efficient supply of ecosystem services by identifying lands that are most likely to provide the greatest return on restoration investments (Barnett et al. 2016). Another important consideration in assessing trade-offs among different ecosystem services involves pathways of co-production, defined by how stages of ecosystem service supply are mediated by human labor, technology, financial capital and institutions (Palomo et al. 2016). Co-production pathways affect the quality, quantity, and distribution of ecosystem service benefits (Palomo et al. 2016). Likewise, analysis of trade-offs and synergies may vary, depending on whether they are based on information regarding supply of ecosystem services, as opposed to their economic valuation (Farber et al. 2016; Turner et al. 2016).

In systematic restoration planning, the key is to minimize potential trade-offs and maximize potential synergies in restoration outcomes. A decision support tool called ROOT (Restoration Opportunities Optimization Tool1) has been developed to help decision makers visualize where financial investments in restoration could be made that would optimize benefits for multiple landscape

1 https://forums.naturalcapitalproject.org/index.php?p=/discussion/833/restoration-opportunities-optimization-tool-root-release

restoration goals (Table 5). Before running ROOT, a preliminary ecosystem services analysis needs to be conducted using InVEST (or other abovementioned models) to estimate the potential impact of restoration interventions. The tool uses linear programming optimization, a method to achieve the best outcome (such as maximum profit or lowest cost) based on a mathematical model whose requirements are represented by linear relationships. The model outputs a suite of optimized restoration allocations, based on different prioritizations of each ecosystem service objective (e.g. reducing sedimentation to rivers or enhancing pollination visitation), generates a map that combines these allocations, and various tables summarizing potential ecosystem service benefits for each allocation decision. ROOT also outputs trade-off curves (efficiency frontiers) that depict the relationship between two alternative restoration objectives to help users identify their optimal restoration strategy.

Efficiency frontiers (also called production possibility frontiers) are a widely used approach to visualizing trade-offs and synergies in systematic landscape planning for conservation or restoration. An efficiency frontier shows the maximum amount of a given objective (e.g. water supply) that can be attained on the landscape for a fixed level of the other objective (e.g. carbon storage) and a given program budget. When the cost of achieving one ecosystem service objective is plotted against another, the efficiency frontier showed all of the possible combinations that fit a given budget. All points along this curve have the same economic efficiency, but the outcomes are different with regard to the ecosystem services provided. The resulting relationship is often a convex curve, but other shapes are possible depending on the specific interactions between ecosystem services within a given landscape (Figure 8). Each point on the frontier shows an optimized restoration allocation scenario. Point A shows a scenario that maximizes potential improvements in water quality, whereas point C showed a scenario that maximizes potential increases in carbon storage. Point B represents an allocation scenario that balances improvements in both carbon storage and water quality. Point D illustrates a non-targeted scenario that fails to increase either ecosystem service. Decision makers can use this analysis to determine where along the efficiency frontier they will target their policies and investments.

https://forums.naturalcapitalproject.org/index.php?p=/discussion/833/restoration-opportunities-optimization-tool-root-release



https://en.wikipedia.org/wiki/Mathematical_model

https://en.wikipedia.org/wiki/Mathematical_model

https://en.wikipedia.org/wiki/Linear_function#As_a_polynomial_function


Efficiency frontiers were modeled in Uganda as part of a multi-criteria spatial optimization study for systematic restoration planning (Gourevitch et al. 2016). Different regions of the country vary in their potential for improving the supply of specific ecosystem services and biodiversity. The efficiency frontier showed clear trade-offs between restoration allocation scenarios that increase aboveground carbon storage and those that improve water quality (Figure 8). This result is explained by the fact that conditions where water quality is improved most differ from those where carbon storage is maximized. Water quality is improved most in areas close to stream networks that have steep slopes and large upslope watershed areas, whereas increase in carbon storage is highest in severely degraded areas with relatively high mean temperatures and annual precipitation and lower seasonality (Gourevitch et al. 2016). Efficiency frontiers with other combinations of objectives, such as carbon storage and biodiversity, water quality and biodiversity, or avoided opportunity costs and carbon storage showed that regardless of where restoration is implemented, carbon storage, water quality and biodiversity will be improved. In particular, restoration of a large area can potentially hold high values for both biodiversity and carbon storage (Gourevitch et al. 2016). When these results were presented to the Ugandan government, the use of efficiency frontiers enabled

decision makers to clearly visualize trade-offs that may influence target areas for investments in restoration. In fact, policy discussions in Uganda led to a preliminary identification of priority districts for FLR.

3.3 Tools for incorporating biodiversity into systematic restoration planning

Biodiversity conservation is an integral component of forest and landscape restoration. But few systematic tools are available for explicitly incorporating biodiversity in restoration planning, particularly for landscape restoration efforts within agricultural landscapes with low levels of remaining forest cover, or in heavily degraded lands. The United Nations Convention on Biological Diversity’s Aichi Targets 5, 14 and 15 motivated the Bonn Challenge and the setting of global targets for FLR (Chapter 1). Nature’s contributions to people ultimately depend on the species that compose ecosystems, the functional properties they possess and the network of species interactions that they create. Biodiversity encompasses multiple levels of variation, including within-species genetic variation, among-species variation, and variation in species composition across habitats and environments.

Figure 8. Efficiency frontiers with optimal restoration allocation maps.

Source: Gourevitch et al. (2016, Figure 2).


Decision support tools for biodiversity conservation focus on selecting priority areas for protection or reserve establishment, with a focus on intact forest areas or large habitat patches. Habitat quality is often used as a proxy for biodiversity using InVEST’s habitat quality model, which estimates the extent of vegetation in a landscape that has reduced threats from highways, built-up areas, and cultivated lands (Baral et al. 2014; Duarte et al. 2016; Lin et al. 2017). But habitat quality models have yet to be validated across the wide range of conditions over which they are applied. The conservation value of a patch of habitat depends on many factors, including surrounding land use and suitability for rare or threatened species (Baral et al. 2014). In fragmented and deforested landscapes, conservation planning focuses on protecting areas that facilitate landscape connectivity, whereas restoration planning focuses on reestablishing connectivity in areas where species movement has been impeded (McRae et al. 2012). Modeling approaches can provide decision support for prioritizing areas within landscapes where restoration will likely bring high biodiversity benefits at reduced cost (Tambosi et al. 2014; Rappaport et al. 2015). These benefits include enhancing landscape connectivity, increasing habitat availability and improving habitat quality or suitability.

Different types of spatially based biodiversity information can and should be incorporated into systematic restoration planning at a range of spatial scales, but there are no tools that provide broadly applicable models, as is the case for ecosystem services. At a national or regional scale, inventory data on the geographic distribution and abundance of species can be used to indicate priority areas for reserves and biological corridors based on criteria such as local species richness, richness of endemic or rare species, refugia for endangered species or assessing species complementarity in protected area networks (Tobón et al. 2017). At the landscape scale, biodiversity assessment data can play an integral role in FLR planning, including prioritizing locations and types of biological corridors, assessing species complementarity networks, and evaluating potential outcomes of different restoration interventions on local- and landscape-scale species and functional diversity. Alternative scenarios can be developed to assess the importance of restoration for conservation of endemic flora and fauna, and to evaluate trade-offs between outcomes that target biodiversity conservation along with a range of other restoration objectives and productive land uses.

As species do not benefit uniformly from restoration interventions, a common approach is to design habitat protection and restoration of landscape connectivity based on an umbrella species concept (Breckheimer et al. 2014). This concept holds that conservation of sufficient habitat for an animal with a very large home range should require areas large enough to provide sufficient area for most species that have smaller home ranges. Considering the potential for species movement within a landscape matrix, an umbrella species is one for which conservation or restoration of its dispersal habitat also facilitates dispersal of other target species (Breckheimer et al. 2014).

An alternative approach for prioritizing areas for forest restoration with biodiversity objectives within human-modified landscapes is to identify which landscapes are likely to offer the greatest overall benefits for biodiversity conservation. The ability of a landscape to support forest-requiring biodiversity is a function of the amount of forest present, the extent of landscape connectivity (Tambosi et al. 2014) and recent forest cover changes (Rappaport et al. 2015). This approach requires information on forest cover thresholds for the persistence of species and taxa within particular regions (Banks-Leite et al. 2014). For example, within the Atlantic Forest region of Brazil, the similarity of species composition of mammals, birds and amphibians in fragmented landscapes compared with intact forest landscapes sharply declines when forest cover falls below 24–33% (Banks-Leite et al. 2014). These threshold relationships can be used to prioritize landscapes where biodiversity has a high chance of recovery, is highly vulnerable due to recent forest loss and where restoration will be less costly.

Several complexities challenge the development and application of decision support tools for incorporating biodiversity in restoration planning at landscape scales. Apart from the importance of overall forest cover, species persistence may also depend on the quality and physical arrangement of forest cover and tree cover within landscapes (Arroyo-Rodriguez et al. 2017). Agroforests, silvopastoral systems, single-species tree plantations, multi-species plantations and naturally regenerating forests vary widely in the number and types of plant and animal species they support (Barlow et al. 2007; Scales and Marsden 2008; Gardner et al. 2009; Chazdon 2014). The spatial scale of landscape effects, as well as the


history of local and regional land use, are likely to be important predictors of the influence of restoration interventions on biodiversity recovery and persistence.

Two studies have used species distribution models to identify restoration priority areas at national scales. Yoshioka et al. (2014) identified priority restoration areas in Japan based on past and current distributions of breeding bird species regarded as endangered, the total area targeted for restoration and the feasibility of restoration. Tobón et al. (2017) used a six-step approach to prioritize potential sizes for restoration that support national scale biodiversity conservation goals in the megadiverse country of Mexico. Their highly robust analysis was based on species distribution models for over 80% of the described species of Mexican amphibians, reptiles, birds and mammals, distribution models of threatened plant and tree species, plant family occurrence data, and locations of critical vegetation types. Two major

sets of criteria, biological importance (priorities for biodiversity conservation, priority vegetation types and presence of secondary vegetation) and restoration feasibility (soil erosion, potential evapotranspiration, elevation zone, land use and fragmentation) were weighted differentially to produce 11 different scenarios. One priority scenario was selected by expert agreement, to produce a map of very high, high and medium priority areas for restoration (Figure 9).

3.4 Assessing and mapping restoration costs and cost-effectiveness

Understanding the factors that influence the financial costs of restoration is central to planning and prioritization, as these costs are highly variable and are often poorly estimated. Iftekhar et al. (2017) describe four major components of ecological restoration costs: acquisition,

Figure 9. Priority sites for restoration in Mexico based on biological importance and restoration feasibility criteria.

Note: Together, priority sites cover 15% of the area of the country’s degraded ecosystems and represent all of the major vegetation types. Existing protected areas are shown in grey.

Source: Tobón et al. (2017, Figure 5).


establishment, maintenance and transaction (Table 6). Currently, no broadly applicable decision support tools are available to estimate and map these costs. Far more work has been done on estimating costs of conservation, which share some commonalities, as opportunity costs may be similar (Naidoo and Ricketts 2006; Armsworth 2014). Wilson et al. (2011) determined restoration implementation cost based on restoration intervention, habitat type and slope. Identifying the factors that influence restoration costs is the first stage of developing decision support tools to estimate them using datasets appropriate for particular restoration contexts (Table 6).

Estimating costs of site-based ecological restoration, however, does not adequately address the full costs of larger-scale landscape, regional or national programs focused on FLR. Scaling up restoration will likely involve economies of scale, and can achieve innovations that increase overall cost-effectiveness and achieve multiple social and environmental benefits that cannot be attained through isolated, local ecological restoration projects (Kennedy et al. 2016). Restoration costs can be highly dependent on the scale of forest restoration activities (Richards et al. 2015). For example, restoration on small properties involves

higher transaction costs than on larger properties. Several studies have evaluated cost-effectiveness scenarios for FLR at local and landscape scales (Macmillan et al. 1998; Birch et al. 2010; Adame et al. 2015; Stefanes et al. 2016; Wada et al. 2017), but much more work needs to be done to scale up these studies to regional or national scales (Nunes et al. 2017).

Across the vast country of Brazil, costs of restoration implementation (establishment) have been carefully estimated. The Instituto Escolhas (2016) created a digital platform to help estimate the financial sums needed to restore forested areas on Brazilian rural properties and the economic returns that forests could generate. The user selects the region to be recovered on the map, inputs the size (in hectares) and the platform offers eight forest recovery models for the user to choose, depending on the area’s degree of degradation and its purpose (permanent preservation areas, legal reserves or only reforestation), with or without financial return. The platform allows property owners or businesses to calculate the monetary sum that needs to be invested by combining different forestry recovery models, with varying portions of each of these models, estimating the value for each of these combinations. This tool was developed

Table 6. Components of ecological restoration costs and factors that can be used to estimate and model these costs.

Type of cost Description Determining factors

Acquisition Cost of acquiring, securing or leasing property to be restored; purchase of permanent or temporary property rights, as for conservation easements, covenants or conservation contracts; opportunity costs or landholder compensation for non-agricultural use of land.

Government policies, discount rates, land quality, land use, market prices, land ownership, financing arrangements, stakeholder engagement.

Establishment Upfront capital investments for implementing restoration, including: engineering works (e.g. in mining sites or wetlands restoration), site preparation, planting or seeding, fencing, weed control and fire control.

Market prices, state of degradation of land, type of restoration, availability and cost of seeds or seedlings, labor costs, financing arrangements, stakeholder engagement.

Maintenance Costs of ongoing management, administration, and monitoring.

Market prices, labor costs, long-term financial, community and institutional support for monitoring and data management stakeholder engagement.

Transaction Costs of selecting sites, organizing programs, negotiating and signing contracts; costs of planning and prioritization.

Government policies, institutional, community and financial support, stakeholder engagement.

Source: based on Armsworth (2014) and Iftekar et al. (2017).


to promote compliance with Brazilian legal obligations for restoration of vegetation on private land (Brancalion et al. 2016b) and uniquely reflects the development of restoration as an economic sector in Brazil.

In the absence of robust data on restoration costs, several types of proxy data are applied to estimate the cost-effectiveness of restoration interventions and as layers in multi-criteria spatial prioritization models (Chapter 4). Land sale, rental cost or gross economic rents for agricultural land are often used as proxies for opportunity costs (Naidoo and Ricketts 2006; Wendland et al. 2010; Silva and Nunes 2017). The potential for natural regeneration – also described as ‘ecological resilience’ (Tambosi et al. 2014; Stefanes et al. 2016) – is a common proxy for implementation costs, as spontaneous forest establishment is far less expensive than full-scale site preparation and tree planting (Birch et al. 2010; Brancalion et al. 2016a; Chazdon and Guariguata 2016; Strassburg et al. 2016). A recent study found that 36% of the obligatory forest restoration debt of 2 million ha in Minas Gerais state, Brazil could be restored using only passive (spontaneous) natural regeneration, at a cost of USD 175 ± 47 million. Adding areas where low-cost assisted natural regeneration would be suitable interventions accounts for 75% of the restoration obligation (1.5 million ha) at a cost of USD 776 ± 137 million. Only the remaining 25% of highly degraded areas require intensive and far more costly tree planting methods (Nunes et al. 2017).

Discount rates strongly influence both cost and benefit estimates of restoration interventions (Ding et al. 2017). Verdone and Seidl (2017) estimated the costs and benefits of restoring 350 million ha of degraded land across the world. Restoration costs were estimated from World Bank reports and the The Economics of Ecosystems and Biodiversity (TEEB) report (Sukhdev 2008). Mean implementation costs per hectare and ranged from USD 389 for lightly degraded areas to USD 3,051 for extremely degraded areas. Restoring 350 million ha of degraded forests produced a net benefit of USD 2.25 trillion at a total cost of USD 299 billion (cost–benefit ratio of 7.54), based

on a 4.3% discount rate and benefits including public and private goods. This analysis highlighted the importance of including the value of public benefits, which lowers the social discount rates, a condition for meeting ambitious global targets for restoration (Verdone and Seidl 2017).


The relevance of models and simulations of restoration outcomes and cost-effectiveness lies in how they are used and interpreted and applied to specific policy contexts within geographic areas. This step may require capacities and skills beyond those of the experts that run the models. Ideally, decision makers should be actively engaged in this stage of the process, as iterative model runs can be used to present a range of scenarios to consider in the process of evaluation (Ruckelshaus et al. 2015). Based on different modeled restoration interventions in different locations, a range of scenarios can be generated to predict the spatial distribution of ecosystem service values, endemic species diversity, or restoration costs. Through an iterative process, relevant scientific information can be incorporated that was not initially known to stakeholders or decision makers.

To be effective as decision support, results and outputs of these tools need to be informed by and interpreted by decision makers who understand how and why they were generated (Ruckelshaus et al. 2015) and who also understand the context in which the model outputs will be applied. As new tools and additional local and global datasets become available, new information can be generated that will be useful in guiding the planning, implementation, and recalibration of the anticipated outcomes of restoration interventions already initiated. Finally, monitoring the outcomes of restoration interventions in terms of ecosystem service supply, biodiversity status or actual implementation costs, provides a mechanism to verify the accuracy and utility of these predictive tools. Participatory approaches for monitoring restoration outcomes and for making corrective actions, if necessary, have been widely applied (reviewed in Evans and Guariguata 2016).

This chapter presents a broad overview of analytical tools to prioritize areas for restoration, and to select restoration approaches and species based on specific objectives and criteria. Tools for guiding the selection of species for planting in particular areas and contexts are also discussed, including approaches that take into account effects of climate change on species distributions. Beyond the spatial prioritization of areas for restoration, additional steps are needed to maximize the success and minimize the cost of restoration interventions within these areas. Costly mistakes can be made by planting poorly adapted or undesirable species in a high-priority area, or by investing in tree planting in areas where assisted or spontaneous natural regeneration are more cost effective (Chazdon 2013; Latawiec et al. 2016; Nunes et al. 2017). Climate change may alter the locations of suitable areas for restoration of particular high conservation value tree species. In some cases, such as pine forests in the western USA, restoration involves thinning trees and setting prescribed burns in even-aged, highly stocked stands to reduce the risk of catastrophic fires and to restore historic fire-resilient forests (Ager et al. 2017). In other cases, restoration interventions require removal of invasive species that threaten native forest ecosystems (Lagabrielle et al. 2011). Spatial prioritization tools can be tailored to provide guidance regarding these challenging planning and implementation decisions.

4.1 Systematic restoration planning

Decision makers involved in FLR need to know which criteria and approaches can help them to identify priority areas for specific FLR interventions within a given region. These interventions can include management and protection of existing forest remnants, setting aside areas for nature regeneration, or tree

planting for ecosystem restoration, agroforestry, woodlots or commercial plantations. Addressing this question requires a systematic approach based on sets of spatially explicit social, biophysical and political data.

Prioritizing areas for FLR is a complex endeavor, involving multiple social and environmental criteria that vary spatially and temporally (Chapter 1). One goal of FLR is to protect and enhance biodiversity by increasing or improving habitat availability, restoring habitat quality and increasing landscape connectivity. A second goal is to restore or improve ecosystem functions by stabilizing water flows, improving water quality, reducing sedimentation and erosion, and storing carbon above and below the ground. A third goal is to improve the well-being and livelihoods of resident communities, by providing employment opportunities through sustainably managing forests for timber and non-timber products, creating locally managed ecotourism businesses and tree nurseries, and by increasing food security, through diversifying and improving local crop production, agroforestry systems and sustainable agricultural practices.

Once these goals are defined within specific geographic contexts (Chapter 2), the question arises: how can we target areas for achieving restoration goals in a beneficial, long-lasting and cost-effective way? Similar complexities and challenges confront conservation practitioners who strive to maximize cost-effectiveness when making decisions regarding allocation of limited funds to protect habitats for endangered or threatened species. The systematic conservation planning approach has been widely used to identify areas of high conservation value based on concepts of representation (Crossman and Bryan 2006; Tobón et al. 2017), complementarity of biodiversity (Ikin et al. 2016) and landscape ecology principles (Margules and Pressey 2000;

4 Tools for prioritization, spatial planning and species selection


Kukkala and Moilanen 2013). This systematic approach to spatial prioritization is now being widely adapted to identify restoration priority areas, using a variety of software and mapping tools (Table 5, Appendix 1).

Approaches to systematic restoration planning first appeared around 2009. Thomson et al. (2009) applied a reserve selection framework to a complex, large-scale restoration-planning challenge, with the goal of restoring landscape connectivity for bird species in a 11,000 km2 rural region of central Victoria in southeastern Australia, where only 14% of the original box-ironbark forest remained. Conservation-planning software Zonation (Moilanen and Kujala 2008) was used to produce a spatially explicit revegetation schedule of priority areas for revegetation, designed to balance habitat requirements for 62 different bird species, based on models of species distributions and projected spatial and temporal models of habitat recovery. In this case, suitable areas for bird habitat had been lost from the landscape and needed to be recreated. Zonation simulated a completely revegetated landscape and then iteratively ‘cleared’ candidate cells until reaching the current extent of native vegetation. Each cleared cell was ranked according to its contribution to future biodiversity value given the current landscape configuration. The most highly ranked cells reflect high-priority areas for revegetation. This example of a single-criterion spatial prioritization model did not consider costs of revegetation or other ecological or social criteria for prioritizing revegetation areas. If sufficient spatial data are available for costs and other prioritization criteria, however, this information can be included within a multi-criteria spatial prioritization framework.

Why use a systematic approach for restoration planning? Priority setting can be accomplished in many ways, including surveys of expert opinion or particular stakeholder groups to provide scores or rankings of priorities (Uribe et al. 2014). Certainly, local communities and government authorities can make decisions about where and how to restore based on their own criteria, objectives and available resources without the need for software or analytical tools. Prioritization decisions become more challenging and contentious when different stakeholders have competing objectives or

trade-offs and consensus is not easily reached. In these cases, restoration interventions may have a low likelihood of long-term success and may result in the loss of time, effort, money and public trust. The value of a systematic approach to prioritization is that the methodological approaches are transparent, and emerge from clearly stated objectives. A systematic approach can also be used to identify potential land-use conflicts, evaluate trade-offs among benefits and assess divergent stakeholder needs by selectively using different sets of criteria in prioritization scenarios.

Systematic approaches are neither top down nor bottom up; they are built on optimization principles that can reflect a wide range of social, political, economic and ecological dimensions (Bryan et al. 2004). For example, Budiharta et al. (2016) tested four restoration scenarios to examine how different evaluation criteria for spatial prioritization of agroforestry interventions on degraded forestlands in East Kalimantan, Indonesia, generate trade-offs in social and ecological benefits for local communities. The results provide guidance for decision makers by comparing outcomes of alternative scenarios, based on different combinations of initial constraints.

4.2 Spatial prioritization tools and their application to restoration

From the perspective of FLR, the main objective of spatial prioritization is to select sites or larger areas for investments and interventions that (1) maximize the long-term success of restoration; (2) ensure representation of biological, geographic and human diversity; and (3) maximize socioecological benefits for local communities and other stakeholders. Given limited resources, the best returns on investments combine feasibility (actions that are practical and possible) with high levels of multiple benefits.

Different types of spatial data sets can provide indicators or proxies for feasibility, and specific restoration benefits. Feasibility can often be reflected in land cost, with areas of low land cost reflecting low opportunity cost and indicating a higher feasibility for restoration. Feasibility also has a social component, as restoration is


more practical in areas where local communities or municipalities are strongly committed and where secure land tenure ensures that landowners benefit economically from restoration outcomes in the long term. The likelihood of long-term restoration success (low likelihood of fire, for example) can also be used as a proxy for feasibility (Wilson et al. 2011). Restoration benefits can be assessed and mapped based on data on remnant vegetation cover, ecosystem services (Chapter 3), habitat improvement, landscape connectivity, livelihoods and farm income, ecological resilience or reduced risk of species loss.

Restoration costs, which are also spatially explicit, are often used as indicators of restoration feasibility (Chapter 3). Restoration feasibility and benefits can also be assessed in terms of their level of urgency, as landscapes with recently elevated levels of degradation or deforestation present a narrow window of opportunity, where low-cost

restoration through natural regeneration remains feasible and is likely to provide multiple benefits (Rappaport et al. 2015).

Spatial prioritization tools can be an integral part of a ROAM process (Chapter 2) or can be applied to optimize restoration interventions within particular management areas (Orsi and Geneletti 2010). During the recent ROAM process conducted in Malawi (Ministry of Natural Resources 2017), spatial multi-criteria analyses were performed to help prioritize investment in FLR interventions, to achieve objectives of food security, resilience to climatic extremes, and biodiversity conservation, based on an underlying spatial assessment of functional degradation of forest and agricultural land. Each of these objectives was assessed and mapped independently based on a large number of input data sets; they were then superimposed to reveal areas where high priorities overlapped (Figure 10).

Figure 10. A preliminary analysis of priority areas identified through multi-criteria analysis.

Note: This figure shows each priority layer and the sum of all of the landscape restoration scenario multi-criteria analyses. Areas in red are the highest priority, whereas areas in blue are the lowest priority.

Source: Ministry of Natural Resources, Energy and Mining - Malawi (2017), Figure 24.


Spatial prioritization analyses for restoration planning can be conducted at a wide range of spatial scales, and can be adapted to specific geographic or political contexts (Table 7). To ensure representation of particular biomes or political units, it may be necessary to conduct separate prioritization analyses within these units. Decision makers can use multi-criteria spatial prioritization to inform restoration planning and investment decisions based on a range of optimization and weighting scenarios. One common application of spatial prioritization tools is to optimize areas for restoration with a high potential for carbon-stocking and co-benefits for biodiversity at a reduced cost (Greve et al. 2013; Renwick et al. 2014; Carwardine et al. 2015). Using a multi-criteria approach at a continental scale, Greve et al. (2013) identified areas of the African tropics where high carbon-stock potential overlaps with high levels of degradation, offering co-benefits of restoration for both biodiversity and carbon storage.

Different types of spatially based data can be used to map and select areas for restoration that satisfy multiple FLR objectives. Using spatial prioritization approaches, multiple criteria can be used to determine the locations that optimize or maximize a set of different functions (Figure 10).

For example, the potential benefits of restoration in terms of carbon stocking, improved water quality or biodiversity conservation do not always coincide spatially, nor do these areas always overlap with areas of high feasibility (low implementation or opportunity costs, high social acceptance of restoration). Gourevitch et al. (2016) mapped the potential of restoration to increase carbon storage, improve water quality and increase biodiversity across the entire country of Uganda, to determine optimal restoration scenarios based on efficiency frontiers for pairwise combinations of these objectives (Chapter 3). Their approach using ecosystem service mapping tools (Chapter 3) and a multi-objective spatial optimization technique enables decision makers to visualize areas within the country with high potential benefits of restoration, where potential trade-offs among these benefits occur, and how the opportunity costs of restoration may affect feasibility (Figure 11). Socioeconomic objectives can also be mapped across the landscape and incorporated into spatial prioritization models (Gimona and van der Horst 2007).

Identifying multifunctional ‘hotspots’ presents opportunities to pursue restoration through coalitions of stakeholders in areas where restoration is likely to offer significant benefits to multiple groups (Gimona and van der Horst 2007). Schulz and Schröder (2017) used a spatial multicriteria analysis to identify multifunctional hotspots in a dry forest landscape in Central Chile, where multiple ecosystem functions (habitat connectivity, erosion prevention and carbon storage) can be enhanced by restoration and where restoration is highly feasible (high natural regeneration potential). The areas designated as multifunctional hotspots provide a starting point for implementation decisions by local communities, non-government organizations and government agencies working within this region.

These and other studies listed in Table 7 demonstrate a wide range of spatial units, prioritization objectives, prioritization criteria and spatial data sets. In some cases, experts were consulted to select appropriate prioritization criteria and indicators (Orsi et al. 2011; Uribe et al. 2014; Tobón et al. 2017). Criteria for prioritization are selected to reflect the overall objectives of prioritization, as well as the constraints on available spatial data sets. Weightings of component factors can be adjusted to examine different scenarios, providing critical information for decision makers. Several studies use only biophysical criteria (biodiversity, vegetation cover, natural regeneration potential, ecosystem functions) to identify priority or target areas for restoration (Gimona and van der Horst 2007; Zhou et al. 2008; Trabucchi et al. 2014; Rappaport et al. 2015; Schulz and Schröder 2017), whereas others include economic criteria such as restoration implementation and opportunity costs, or farm profitability (Crossman and Bryan 2009; Lagabrielle et al. 2011; Wilson et al. 2011; Crouzeilles et al. 2015; Zweiner et al. 2017). Three studies emphasize social feasibility and biophysical suitability criteria (Orsi and Geneletti 2010; Uribe et al. 2014; Vogler et al. 2015) and one study includes political feasibility, based on government restrictions on land use for agroforestry in production forests (Budiharta et al. 2016). The study by Tobón et al. (2017) identifies priority sites for restoration within the entire country of Mexico, to achieve the Convention on Biological Diversity Aichi Target 15 to restore 15% of terrestrial ecosystems at minimal cost. In this case, criteria for prioritization were biological importance and restoration feasibility. Stefanes et al. (2016) use multi-criteria analysis to compare three restoration scenarios based on different restoration targets in the Cerrado biome of Mato Grosso, Brazil.

Decision support tools for forest landscape restoration

|35

Table 7. Spatial prioritization studies for restoration planning.

Authors Geographic region

Size of study area Objective Optimization criteria Datasets used

Gimona and van der Horst (2007)

NE Scotland No data Identify suitable target areas for creation of small multifunctional woodlands.

Biodiversity, visual amenity, recreation.

Spatially explicit habitat suitability models for 16 species deemed to be priority species in a Biodiversity Action Plan.

Zhou et al. (2008)

Upper Yangtzee River Basin, China

7,400 km2 Identify potential restoration methods and priority areas for increasing forest cover from 21 to 30%.

Areas of shrublands that were former forest that are adjacent to existing forest fragments.

Vegetation maps.

Crossman and Bryan (2009)

Murray-Darling Basin, Australia

11.87 M ha Identify hotspots for restoring natural capital and increase vegetation cover by 1%.

Maximize areas of greatest benefit to restoring natural capital with minimal impact on farm income.

Maps of bioregions, climate zones, soil classes, remnant vegetation, probability of soil and wind erosion, deep drainage rates, depth to groundwater, annual rates of carbon sequestration, farm profitability.

Orsi and Geneletti (2010)

Western Chiapas, Mexico

18,500 km2 Identify reforestation areas important for biodiversity conservation that have high feasibility (restorability).

High biodiversity value (multiple criteria) and high feasibility (multiple criteria); additional ecological and social criteria also included).

Land-cover maps, risk of soil erosion, topography, distance from urban areas, land-use conversion cost, aggregated poverty index.

Lagabrielle et al. (2011)

Réunion Island

2,512 km2 Identify the best spatial options for conserving and restoring a representative sample of biodiversity features and ecosystem processes, while minimizing conflicts with other land uses.

Achieve conservation and restoration targets with minimal costs, including restoration and management costs.

Map of 44 habitat classes, map of spatial components of biodiversity processes, map of 23 natural corridors, map of existing reserve network, and conservation costs index.

Orsi et al. (2011) Central Chiapas, Mexico

430 km2 Identify reforestation priorities that permit both timber production and biodiversity protection.

Maximize reforestation for timber harvest and for conservation, minimize ecological cost.

Modeled potential species richness, forest cover change maps, plantation and fencing cost, maps of regeneration potential.

Wilson et al. (2011)

Irvine Ranch National Landmark, California, United States

1353 ha Identify the combination of restoration sites and the schedule for their implementation most likely to deliver the greatest utility for a fixed budget and operational constraints.

Maximize the utility of restoration, accounting for the costs of restoration, the likelihood of restoration success, the probability of stochastic events, feedbacks, time lags and spatial connectivity.

Potential and actual vegetation maps, slope and aspect (used for estimating restoration cost and likelihood of restoration success).

continued on next page

| Robin L C

hazdon and Manuel R G

uariguata36



Greve et al. (2013)

African tropics 3.6 million km2 Identify areas most suitable for carbon sequestration projects while also benefitting biodiversity, groundwater recharge, land value and governance.

Areas with maximum co-benefits given different weightings of carbon-stock potential.

Maps of aboveground carbon storage potential, distribution maps for all species of mammals, reptiles and amphibians, groundwater recharge, land cost, Ibrahim Index of African Governance.

Trabucci et al. (2014)

Martín River Basin, NE Spain

2,112 km2 Identify priority areas for restoration in a degraded opencast mining area in a semi-arid Mediterranean river.

Maximize provision of multiple ecosystem services.

Combination of erosion data (level of degradation) with five ecosystem services maps: erosion control, maintenance of soil fertility, surface water supply, water regulation and carbon storage in woody vegetation.

Uribe et al. (2014)

Upper Mixtec region of Oaxaca, Mexico

8,100 km2 Identify priority areas for FLR that considers environmental and socioeconomic criteria and evaluates them according to the opinion of different participant groups.

Maximize the consensus of stakeholder criteria and priorities.

Maps of land suitability were generated, reflecting the preferences of different stakeholders. An overall map of FLR priorities was developed by combining the priority maps of the four groups.

Carwardine et al. (2015)

Australia, entire continent

Heavily cleared vegetation suitable for carbon forestry

Identify priority areas for biodiverse carbon plantings that most cost-effectively restore heavily cleared vegetation types to 30% of original extent while sequestering carbon.

Priority areas that meeting 30% vegetation restoration targets at a minimal opportunity cost, without compromising carbon sequestration.

Map of 139 vegetation types that have been cleared to below 30% of original extent, aboveground biomass for 53 sites, opportunity cost (value of agricultural land), carbon selling price (net transaction costs), restoration cost.

Crouzeilles et al. (2015)

Atlantic Forest of Brazil

1.5 M km2 Identify planning units for two mammal species that, if restored, would cost-effectively improve population dynamics in this fragmented biome, while also preserving any broad-scale genetic and phenotypic diversity that may exist.

Planning units that achieve specific restoration targets while minimizing cost.

Species distribution map and data on dispersal and habitat availability; data on land use, land cost and distribution of forest remnants.

Table 7. Continued


Decision support tools for forest landscape restoration

|37



Rappaport et al. (2015)

Atlantic Forest biome in southern Bahia, Brazil

2 million ha Identify priority landscapes for restoration that maximize contributions to regional-scale connectivity at lowest cost, based on the urgency and feasibility of interventions.

Maximize restoration outcomes for biodiversity in landscape units with intermediate forest cover, recent landscape change, and high potential for regional connectivity.

Across landscape units, calculate indicators of regional importance (regional connectivity), restoration feasibility (forest cover percentage), and urgency (recent change in landscape structure).

Vogler et al. (2015)

Eastern Oregon, United States

914,657 ha of national forest

Analyze priorities and trade-offs among a range of socioecological objectives affecting ecosystem services in a large fire-prone national forest.

High-priority planning areas to target restoration efforts with respect to the attainment of multipleobjectives (restoration of forest stressors vs. producing provisional ecosystem services (e.g. timber).

Vegetation departure from historical reference (restoration need), insect risk, timber volume, wildfire risk to the wildland urban interface (WUI), and potential wildfire hazard.

Budiharta et al. (2016)

Paser district, East Kalimantan, Indonesia

1.16 million ha Prioritize restoration of degraded lands using agroforestry systems that provide livelihoods and subsistence for local communities.

Maximize social feasibility and biophysical suitability for agroforestry while minimizing opportunity cost (profits foregone from alternative land uses for oil palm or timber harvest).

Biophysical suitability, compatible land cover types, map of forest concessions, economic and cultural dependency on agroforestry, preference for agroforestry over other land uses.

Fernández and Morales (2016)

Shrub and sclerophyllous forest ecological region of Central Chile

No data Identify priority areas for restoration of two threatened endemic tree species, and analyze how different environmental conditions and land-use types may affect the selection of areas for species restoration under climate change.

Maximize environmental conditions suitable for each species and land types suitable for restoration initiatives, accounting for future climate change.

Environmental suitability (current and future potential species distribution) and land suitability variables (land-use type and priority sites for conservation).

Gourevitch et al. (2016)

Uganda Entire country Allocating the optimal amount of land for restoration within each district.

Maximize the value of benefits from carbon storage, water quality and biodiversity, and minimize opportunity costs.

Forest cover lost between 2000–2013; potential aboveground carbon storage; soil erosion map based on existing and restored vegetation; species richness maps of mammals, reptiles, amphibians, and birds, value of annual agricultural production (proxy for opportunity costs).


Table 7. Continued

| Robin L C

hazdon and Manuel R G

uariguata38



Stefanes et al. (2016)

Cerrado region of Mato Grosso do Sul state, Brazil

216,086 km² Compare three restoration scenarios based on different restoration targets.

Maximize restoration benefits and minimize opportunity costs from foregoing agricultural land use.

Native vegetation cover, landscape connectivity and agricultural productivity in 10,000 ha planning units.

Schulz and Shröder (2017)

Central Chile 13,175 km2 Identify potentially feasible forest restoration areas that simultaneously contribute to the enhancement of multiple forest functions (multifunctional hotspots).

Maximize restoration feasibility and multifunctionality.

Maps of potential erosion prevention, potential carbon sequestration, and potential habitat function and connectivity.

Tobón et al. (2017)

Mexico 60% of terrestrial land area of Mexico

Prioritize potential sites for restoration to support conservation and recovery of biodiversity to achieve 15% of Mexico’s land area to address Aichi Target 15.

Maximize biological importance and restoration feasibility (relative restoration cost and likelihood of success).

18 spatial datasets based on criteria of Orsi et al. (2011b).

Ager et al. (2017) Western United States

245,000 ha in Deschutes National Forest, Oregon

Prioritize project areas that most efficiently reduce potential wildfire loss of fire-resilient ponderosa pine stands, while creating contiguous areas within which prescribed and managed fire can be effectively used to maintain low-hazard conditions.

Optimize the location of projects to facilitate the efficient future use of prescribed and natural fire as a means to sustain fire resilient forests and reduce the likelihood of uncharacteristic fire and associated ecological loss.

Spatial variation in fuels, potential fire behavior and tree density.

Zweiner et al. 2017

Brazilian Atlantic Forest

1.5 M km2 Identify priority sites for conservation and restoration of woody species to achieve conservation targets of 10%, 17% and 20% of the Atlantic Forest extent.

Maximize conservation of plant species under contrasting climate and land-use scenarios, considering socioeconomic costs, presence of protected areas and distribution of existing forest remnants.

Species ecological niche models, climate data under two scenario projections, land-use data, land acquisition cost, map of remaining forest areas, map of strictly protected areas > 1000 ha, map of percent of GDP invested in environmental programs (municipality level).

Note: Not all are multi-criteria studies. Studies used a range of different software applications and modeling approaches. See author citations for more details.

Table 7. Continued


Because forest restoration and recovery of ecosystem functions are spatially and temporally dependent processes, modeling scenarios of restoration should incorporate these dynamics (McBride et al. 2010; Rappaport et al. 2015), as well as the effects of climate change on habitat suitability. The restoration prioritization approach of Wilson et al. (2011) incorporated both spatial and temporal variation, accounting for the probability of stochastic events, feedbacks, time lags and spatial connectivity. Several prioritization analyses take into account the likely changes in species distributions due to climate change.

Fernández and Morales (2016) identified priority areas for the restoration of two threatened endemic tree species in central Chile, based on environmental conditions corresponding to likely climate change scenarios and land-use types that provide suitable areas for restoration (Table 7). In Brazil’s Atlantic Forest region, Zweiner et al. (2017) identified priority sites for the conservation and restoration of woody species under two climate change scenarios. In these cases, the modeled benefit of restoration is a reduced risk of extinction of endemic species due to climate change.

Figure 11. Maps of the potential for restoration in Uganda to increase carbon storage (A), improve water quality (B), increase biodiversity (C) and incur opportunity costs (D).

Note: Areas where restoration has a high potential impact are in dark green, whereas areas showing a low potential impact of restoration are in dark purple. Values are standardized (rescaled) for each map, where zero is the mean value.

Source: Gourevitch et al. (2016).


4.3 Spatial tools for prioritizing areas for specific restoration interventions

Another use of spatial prioritization tools is to identify priority areas for implementing specific types of restoration interventions, such as natural regeneration, agroforestry or sites for prescribed burning. For restoration objectives that focus on biodiversity conservation, multiple ecosystem services and cost-effectiveness, natural regeneration should be prioritized, but not all areas of deforested or degraded land are suitable for natural regeneration, due to soil disturbance or poor seed dispersal from areas of remnant forests. Identifying and prioritizing areas with a high potential for natural regeneration of forests can greatly increase the cost-effectiveness of restoration within landscapes (Chazdon and Guariguata 2016; Nunes et al. 2017). Several promising regional approaches for identifying priority areas for natural regeneration could be expanded and replicated in other regions and countries. Zhou et al. (2008) identified areas in Upper Yangtze River Basin where natural regeneration is likely to be a successful forest restoration approach using soil taxonomy to predict potential vegetation type and identify potential tree species for restoration and restoration methods to achieve 30% forest cover. The priority areas for rapid restoration are former forests that are now covered by shrubs but are adjacent to existing forestlands, and are predicted to have an adequate supply of seed from local tree species.

To examine how low-cost restoration pathways can help landowners comply with legally mandated restoration under Brazil’s Native Vegetation Protection Law, Nunes et al. (2017) mapped the potential for natural forest generation on existing pasturelands in the Brazilian state of Minas Gerais (Table 7). Prioritizing areas of medium to high natural regeneration potential results in high cost savings, and also produces high benefits in terms of carbon storage, water quality and biodiversity (Nunes et al. 2017).

Budiharta et al. (2016) focused their analysis of restoration prioritization in Indonesia on local tree-based agroforestry, which provides livelihoods for local communities and contributes to food security, while offering multiple ecological benefits of tree cover. The agroforestry systems used in this region over the past 300 years are commonly established through enrichment planting in degraded forests

or young secondary forests, using seedlings and wildlings of various tree species. Their analysis was based on 11 biophysical requirements of the most important tree species used in these agroforestry systems, climate and rainfall data, land cover maps indicating compatibility with agroforestry land uses, and data on social and political feasibility. Analysis of four different scenarios provided insights into the location of priority areas for agroforestry, given different combinations of biological, social and political constraints (Figure 12).

Forest restoration on public lands in the western United States focuses on restoring fire-prone ecosystems to fire-resilient ecosystems with a reduced risk of severe and extensive wildfire. Logging practices, grazing and fire suppression policies have impacted much of the conifer forests in the western United States, leaving them highly vulnerable to severe wildfires. Ager et al. (2017) identified priority areas for management interventions in central Oregon, such as thinning, introducing low severity ground fires and seeding of fire-adapted native grasses and shrubs.

4.4 Spatial tools for selecting appropriate species for agroforestry and restoration

Without proper planning or species information, selection of tree species for restoration plantings, woodlots, protection forest or agroforestry can lead to high rates of tree mortality or other problems, such as poor economic returns or adverse environmental consequences. In the worst cases, restoration projects end up causing increasing levels of degradation, rather than decreasing and reversing effects of degradation. The costs can be high, and include loss of trust and cooperation with local communities. Several decision support tools have been developed to provide guidance regarding the right tree to plant for the right place and purpose. As with other restoration decisions, local stakeholder engagement is critically important, as the particular tree species planted influence local livelihoods, incomes, food and timber products and management protocols.

An online species selection tool developed by Kindt et al. (2015a) is available for Google Earth, as part of the Vegetationmap4africa web-based map. The species selection tool provides the potential natural vegetation (PNV) map as Google Earth


layer, overlaid by point layers with ‘clickable labels’ that approximate the distribution of vegetation types for a particular country. The tool provides information about tree and shrub species that have been shown to be useful to farming or pastoral communities throughout eastern Africa. The

online tool also provides additional information for species selection that covers the whole African continent, but at a lower spatial resolution. The tool aims to help in the selection of useful tree species for planting anywhere in Africa. A related tool developed by the International Center for

Figure 12. Maps of priority areas for agroforestry restoration under four social-ecological scenarios: (a) Scenario 1 (ecological variables only); (b) Scenario 2 (ecological and social variables); (c) Scenario 3 (ecological and social variables with existing political constraints); (d) Scenario 4 (ecological and social variables with constraints associated with land use removed).

Source: Budiharta et al. (2016).


Agroforestry, The Agroforestry Species Switchboard, provides access to a wide range of information sources and databases for over 26,000 plant species used in agroforestry and restoration initiatives (Kindt et al. 2015b).

Another new tool using the Google Earth interface allows the user to identify which tree species could possibly be used at a particular restoration site according to their expected ability to persist there under current and future climatic conditions (Thomas 2016). The tool is currently under testing by users. A prototype of the tool has been developed to focus on seasonally dry tropical forest regions of Colombia, a highly threatened ecosystem with more than 430 tree species (http://www.restool.org). The user locates a specific restoration area in Google Earth, and inputs information regarding how many tree species they aim to plant and what life history characteristics are desirable. Then, prevailing environmental and stress conditions at the restoration site need to be input, such as compacted soil, steep slopes or vulnerability to fire. Finally, the restoration objectives are provided. Based on this user-defined information, the tool searches for tree species that are expected to be able to persist at the restoration site under climate change, that will be tolerant of local stress factors, and whose functional trait profiles are best matched to the restoration objectives. In addition, the tool generates maps of recommended areas for seed sourcing to meet the restoration objectives. Users can also request information on recommended propagation methods and management of the species to be included. Bioversity International scientists intend to extend this tool for use in other regions of the world.

Multiple criteria are often used to select tree and shrub species for planting under specific conditions. Reubens et al. (2011) developed a multi-criteria tree species selection tool for prioritizing tree species use in Tigray, northern Ethiopia, that could potentially be used across the East African region (Reubens et al. 2011). The model uses a species-level database, including information on socioeconomic functions, sociocultural values, environmental services, general and local plant preferences of local people, and biodiversity relevance. Species with the highest overall scores can then be selected based on particular environmental and use criteria. Two multiple-criteria approaches were used to prioritize tree, shrub and grass species for gully and riverbank stabilization in southwest Ethiopia (Talema et al. 2017).

A decision support tool has recently been developed for selecting tree species in coffee and cocoa agroforestry systems in Uganda and Ghana, with potential for application in other countries (van der Wolf et al. 2017). The tool helps to select appropriate shade-tolerant tree species given local conditions and the needs and preferences of smallholder farmers, while also providing different ecosystem services in local plots and the surrounding landscape (Figure 13). Users of the tool are extension officers who work with cooperatives and rural development projects. The tool is based on the knowledge and experience of local farmers, who rank tree species according to their services, shade quality, pest and disease suppression ability or effects on soil fertility. One or more ecosystem services can be selected as criteria for selecting species (van der Wolf et al. 2017).

Working in a tropical riparian forest setting, Meli et al. (2014) combined five independent selection criteria to yield a species selection index for riparian restoration: (1) species importance in reference forest; (2) natural regeneration potential following disturbance; (3) habitat breadth; (4) social acceptance; and (5) propagation requirements in terms of time and financial investment (Meli et al. 2014). One advantage of this selection procedure is the independence of the five criteria, and the potential to select different species in projects with different restoration goals. Likewise, local knowledge and perceptions of tree species importance and uses are taken into account in restoration decisions.

Plant functional traits provide useful criteria for selecting tree species for planting under particular conditions and for ensuring that selected species vary in their ecological functions. Measurements of 10 traits for 118 native tree species were used to select priority species for restoring forests in the ironstone outcrops of Serra dos Carajás in Pará, Brazil (Giannini et al. 2016). Traits included seed dispersal mode, pollination, growth characteristics, geographic distribution, endemism, occurrence on outcrops with high levels of iron and uses for local people. Using non-metric multidimensional scaling (NMDS), a subset of species can be selected for use in restoration plantings that maximize functional diversity (high variability) as well as local adaptation and use value. This approach could be easily adapted for tree species in other regions to select groups of local species for restoration plantings that will maximize functional diversity and ecological interactions with local fauna.



Whether the goal is prioritizing species for planting under specific conditions or prioritizing areas for optimizing forest restoration interventions, multi-criteria support tools enable a systematic approach to decision making that is transparent and multidimensional. To be effective, the criteria and data used in these models must be carefully chosen to reflect the main objectives of restoration, and to minimize known trade-offs in outcomes. Prioritization tools can highlight

important synergies and trade-offs that can help decision makers and practitioners determine how and where to achieve the most desirable and feasible restoration outcomes. Even the best tool, however, may fail to overcome spatial resolution limitations of available data. Coarse-scale prioritization results may therefore fail to indicate optimal restoration outcomes within management units, where conditions that vary at small spatial scales ultimately influence restoration outcomes and the performance of different tree species.

Figure 13. A flowchart showing user steps when using a decision support tool for tree selection on smallholders’ coffee farms.

Source: van der Wolf et al. (2017, Figure 2).

5.1 Tools for what and for whom?

Tools are used to build and repair things, but are ineffective without a clear objective in mind and without a plan of action. Using the wrong tool can be counterproductive and wasteful in both time and money. In the case of FLR, the plan can be a framework or a progression of steps and activities to guide a long-term implementation and adaptive management. Prior to embarking on implementation, scenarios can be modeled that indicate where and how outcomes would best fulfill objectives for the lowest cost or with the highest feasibility. Planning and implementation steps are guided by the principles of FLR, as described in Chapter 1. The process begins with stakeholder engagement, assessment of restoration opportunities at various spatial scales, and building consensus regarding restoration objectives (Chapter 2). These activities can be accompanied by the use of support tools and models designed for planning how and where to implement specific restoration interventions, and how to optimize restoration outcomes and cost-effectiveness based on multiple social and environmental criteria (Chapter 3). The information used in spatial prioritization can be based on spatially explicit assessments of the quantity, quality, and value of ecosystem services, and scenarios describing how restoration interventions will impact the conservation and distribution of species of interest or concern (Chapter 4).

An important issue is who is using these decision support tools. The most sophisticated models are generally used by highly trained staff (who are often modelers themselves) who work for organizations, government agencies, or research centers. Some of the most useful tools, however, are those that can be used by practitioners, or even local communities, and can be understood by policy makers. So, if the goal is to reach policy makers, results of relatively simply models are more likely to be adopted, provided they are clearly documented, published

and have been validated to reveal their limitations (Ruckelshaus et al. 2015). In addition, simple open-source tools, such as InVEST, help to facilitate an iterative process because local experts can be trained to use models and supply ongoing technical support. Moreover, the credibility of model information provided to decision makers is greatly enhanced in cases where locally invested scientists have leadership roles in providing input data, or interpreting and translating model outputs (Ruckelshaus et al. 2015).

5.2 The power of visualization

Visualization is a powerful tool for consensus building, negotiation and evaluating future scenarios in a landscape (Boedhihartono 2012). Visualization exercises, such as animated drawings and smart-mapping, can enhance communication and stakeholder engagement, and encourage constructive dialogue with policy makers regarding restoration scenarios. Smart-mapping is a method used for exploring scenarios using computer-generated drawings based on a library of symbols and icons. A variety of landscape scenarios can be created in a more standardized way than hand drawing, making the images more attractive to government officials, private sector representatives and other stakeholders. Smart-mapping can be used as a tool to explore landscape scenarios with policy makers, and to help understand the important drivers of change in a landscape. For example, Figure 14 shows a landscape drawing by villagers in eastern Sudan of the original landscape and two alternative future landscapes, one envisioned by women and one envisioned by men. The women preferred a road linking mango trees to markets, as dried mangos and mango jam contribute to their income. The men put more electrical sources in their alternative landscape, along with more trees in the plantations and a sports center. The smart-mapping exercise helped communities to reach a compromise on an action plan, which was later implemented with assistance from local government and international organizations.

5 Conclusion: Synthesis and outlook


Figure 14. Smart-maps created by villagers in East Sudan.

Source: Boedhihartono (2012).

Present landscape

Women’s vision of the future

Men’s vision of the future


Drawings and ranked assessments can also be used as baselines for monitoring change in ways that record and communicate diverse values of people within the landscape. Drawings can be archived and viewed after several years to illustrate how scenarios have changed, providing rich opportunities for discussion and evaluation (Boedhihartono 2012). Drawings can be supplemented by photographs, videos or smart-mapping. The Landscape Outcome Assessment Methodology (LOAM) has been used by the World Wildlife Fund to measure, monitor and communicate changes in a landscape over time. This approach is based on scoring five assets: natural, human, physical (built), social and financial, which are identified through a participatory process and recorded in radar diagrams. Repeated analysis over time can show how the assets scores and the balances between them are changing (Aldrich and Sayer 2007).

5.3 Theory of change for FLR

Despite the orderly graphs and frameworks, the FLR process is neither tidy nor linear, and requires sufficient momentum, guidance and support from public and private sectors to withstand obstacles, crises, shocks and their complex reverberations across political, economic, social and ecological domains. Clearly, adaptive management, monitoring and long-term active engagement of stakeholders are key to the long-term viability and success of FLR and other landscape approaches (Reed et al. 2017; Sayer et al. 2017). Original objectives that motivated restoration interventions may have to be modified in response to both external and internal drivers, including climate change, natural disasters, legal instruments, migration, financing arrangements and land-tenure policies. Therefore, FLR needs to have a theory of change (Weiss 1997; Sayer et al. 2017). Visualization and conceptualization of FLR are early phases in the process (Stanturf et al. 2017), but over time these steps need to be revisited. The need for flexible and adaptive management is challenging to reconcile with long tree lifespans and the long timescales needed to reach the full benefits of restoration interventions. Indicators of success may need to focus on short-term proxies for long-term metrics. These factors need to be taken into account when planning interventions that can potentially be redirected towards different objectives. A theory

of change defines an expected pathway between an intervention and its ultimate impact, and exposes the assumptions underpinning predicted outcomes. Different indicators and metrics are needed at multiple stages throughout the process to inform progress and decision making (Sayer et al. 2017). Ideally, local residents have sufficient capacity, organization and financial resources to become landscape restoration practitioners who develop the indicators and metrics that will permit adaptive management of restoration interventions and impact pathways.

FLR needs innovative approaches to transform problems into solutions. Short-term financial returns should be built into the early stages of FLR interventions, providing time for benefits for biodiversity and ecosystem service supply to accumulate. One example is the intercropping of short-lived Eucalyptus trees with native species in restoration plantings. Income from harvesting short-rotation Eucalyptus compensates for the costs of planting and maintaining the plantations. Harvesting the rows of Eucalyptus trees then favors the growth of planted and naturally regenerating native tree species (Brancalion and van Melis 2017).

5.4 New frontiers for decision support tools for FLR

Decision support tools for FLR are proliferating rapidly, and it is likely that by the time this report is published more new tools will be developed and put to use. Tools are now being developed to select criteria and indicators for monitoring restoration outcomes at landscape and national scales (Buckingham et al. 2017), and for conducting economic and policy analysis in support of restoration and policy decisions (Ding et al. 2017). Few support tools provide guidance for monitoring restoration outcomes. Participatory monitoring involving local people could meet the needs of monitoring restoration outcomes within and across countries (Evans and Guariguata 2016). This review reveals a gap in decision support tools focused on the implementation of landscape-scale restoration initiatives. Most tools focus on assessing restoration opportunities at a coarse scale within countries, rather than at a finer scale within landscapes, where implementation happens and where the costs and benefits of restoration have direct impacts.


Decision support tools are an important interface between science, policy and practice in the forest landscape restoration arena. As the science of restoration grows, new knowledge can be used to develop new and improved support tools that are more accessible and easy to use. For example, models for predicting the capacity and quality of natural regeneration in Brazil’s Atlantic Forest region are being applied in restoration planning efforts (Chazdon 2017). Similarly, new research on the role of governance, intrinsic motivation and institutional arrangements will inform the development of decision support tools with respect to stakeholder engagement in planning and implementation, and for policy development regarding sharing of the public benefits of restoration within landscapes. On the economic side, more data are needed on the costs of restoration and its spatial determinants, so that restoration planning can incorporate analysis of cost-effectiveness of restoration interventions. Economic analyses need to account for the full benefit of restoration interventions, including non-material and material benefits.

Further developments of ROAM could provide more guidance regarding different modes of restoration and their expected ecological and economic outcomes. Areas that require enhancement with regard to decision support tools are: estimating the dynamics of carbon storage in different restoration scenarios, incorporating biodiversity data in restoration opportunities assessments and developing scenarios for the economic benefits of restoration interventions in relation to food security and local livelihood support.

Despite the many models and tools available in support of systematic restoration planning, social aspects of restoration remain poorly considered compared to biophysical aspects. We know far more about the current or potential supply of ecosystem services than about how they are used by stakeholders, how they support livelihoods, how and which people are involved in co-production, and how ecosystem service flows are influenced by local governance and power relations (Lele and Srinivasan 2013; Paudyal et al. 2016). Mapping of social landscapes shows enormous potential in understanding how decisions are made, and how gender, race and social class influence access, rights and power in land-use decisions central to restoration planning and implementation (Buckingham et al.

2017). Analysis of power networks and relationships can provide critical information for developing agent-based models of landscape change under different restoration scenarios (Baynes et al. 2016).

Participatory restoration and collaborative restoration engage stakeholders in all phases of the restoration process, increasing trust and transparency and resulting in greater likelihood of long-term success. A framework for participatory restoration outlines four major steps where stakeholders can become actively involved: conceptual planning, preparatory activities, implementation, and evaluation and monitoring (Derak et al. 2017; Figure 15). The framework was tested in North Africa, confirming that a well-designed participatory program encouraged stakeholder engagement in forest restoration (Derak et al. 2017).

The Cooperative Forest Landscape Restoration Program was mandated by the US Congress in 2009 to encourage collaborative, science-based ecosystem restoration of priority forest landscapes within national forests (Urgenson et al. 2017). Urgenson et al. (2017) explored the complex socioecological processes involved in developing a shared vision for collaborative restoration of fire-adapted forest landscapes in the western United States. Their study of six collaborative FLR programs showed three broad categories of challenges: meeting multiple objectives, collaborative capacity and trust, and integrating ecological science and social values in decision making. Mandating collaboration can encourage stakeholder engagement and renew commitment to overcome prior conflicts, but can also lead to increased scrutiny and tensions in adjusting to the collaborative process (Monroe and Butler 2016).

Decision support tools rely heavily on land cover data, which is often derived from satellite imagery or aerial photographs. But much qualitative and quantitative information about the fauna and flora associated with these land covers is not provided by these data sources, even if they are measured at high spatial resolution (Dirzo et al. 2014). As a result, we lack important information about the multifunctionality of existing landscapes, and how these functions could be affected by restoration interventions (Turner et al. 2016). Research on the effects of landscape configurations and structure, as well as on the composition of plant and animal species, is critically important for guiding restoration planning based on remote sensing data (Pardini et al. 2010; Banks-Leite et al. 2014).


Degradation is often the main focus of assessing restoration opportunities and interventions, although the threshold for classifying degradation is highly subjective. Degradation is viewed as ‘the problem’, and restoration is ‘the solution’. But restoration is not the goal; rather, it is the means to achieve multiple goals. In reality, the problem is lack of food security, inadequate water supply, high risk of catastrophic fires, climate change and loss of species. Well-planned FLR programs can go a long way to resolve these problems without the need for land degradation assessments. Furthermore, assessments of land degradation can be misleading if they fail to account for trajectories of forest recovery processes (Chazdon et al. 2016; Ghazoul and Chazdon 2017). One stakeholder’s perception of what is degraded may be different from another’s. The ‘false degradation’ narratives can be used to convert secondary forest into reforestation using exotic species monocultures, under the premise that reforestation will improve ecosystem services, provide jobs and improve livelihoods of local people (Hajdu et al. 2016). Hajdu and Fischer (2017) developed a framework for analyzing degradation narratives that can be used to explore the extent to which these are

based on evidence. The framework was applied to degradation narratives in relation to forest carbon interventions, but can be applied to any specific case where degradation-related problems and solutions are cited. The framework is designed to be easy to use and to help practitioners, researchers and policy makers distinguish between genuine cases of environmental degradation and invalid claims of degradation that serve to promote the narrow self-interest of particular actors.

An alternative view to the notion that degradation is the most appropriate index to use when planning restoration interventions is to focus instead on areas where multiple restoration benefits can be achieved for minimal cost. Many of the spatial prioritization approaches discussed in Chapter 4 identified areas where the social and environmental benefits of restoration are likely to be high, and where restoration is also more feasible. For example, natural regeneration is most likely to occur in areas with low soil degradation that are close to existing forest patches, incurring far lower costs than tree planting (Chazdon and Guariguata 2016). In areas with highly degraded soil that are distant from forest patches, full tree

Figure 15. A proposed framework for participatory restoration in rural areas of North Africa.

Source: Derak et al. (2017, Figure 1).

Participatory forest

restoration

1. Conceptual planning:

Discussion of restoration feasibility & objectives Consensus on the decision of restoring

2. Preparatory activities:

Administrative, �nancial & logistical issuesTechnical procedurePeople information & sensitization

3. Implementation:

Voluntary participationRemunerated �eld tasks

4. Evaluation & monitoring:

Collect comments, critics & suggestionsSite supervisionScienti�c monitoring


planting of native tree species is required to restore forest ecosystems. From a landscape perspective, however, these highly degraded areas could be rehabilitated to sustainable croplands, agroforests, woodlots or native species plantations that would provide livelihoods and ecosystem services to local communities and may provide a greater financial return on investment.

5.5 Challenges to operationalizing FLR

The landscape approach has gained traction in the sustainable development agenda as a way to overcome the limitations of sectorial approaches, to address trade-offs within spatial units, and to enable a deeper understanding of change and resilience in socioecological systems (Freeman et al. 2015; Bürgi et al. 2017). But, how do we operationalize a landscape approach such as FLR? The question is better stated by Turner et al. (2016, 202): “Can we envision various realistic and probable scenarios and diligently implement, enforce, and maintain the policies and practices needed to achieve those scenarios we collectively decide are most desirable?” And, how do we know when we are achieving this vision?

One difficulty in operationalizing the conceptual framework of landscape approaches is that ‘landscape’ is a spatial concept rooted in a geographical pattern that is flexibly defined and does not convey a practical meaning of management, control, ownership or policy within this physical space (McCall 2016). McCall (2016, 62) proposes viewing landscapes as ‘territories,’ defined as “a geographic space where a society, and/or political entity, shapes, influences, and controls social activities and access to resources”. Both landscapes and territories are geographic spaces, and they do sometimes coincide, but

landscapes are biophysical entities whereas territories are political and social entities. The very notion of ‘landscape’ does not easily lend itself to an operational form where transactions, ownership, property boundaries, political entitlements and rules govern everyday activities. The landscape concept has ‘constructive ambiguity’, meaning that it is appealing and widely used, but there is no consensus about its meaning in practice (Sayer et al. 2013; Freeman et al. 2015).

Principles and guidelines can point to criteria, indicators and standards of FLR, but so far, we only have the principles and a few scattered guidelines (Brancalion and Chazdon 2017). FLR is a relatively new approach to land management and to restoration, although there are historical antecedents. To promote implementation of FLR on a large scale, and to encourage long-term private and public investments in FLR planning and implementation, some basic and general standards will be needed to ensure that principles are upheld, to reduce risks of failure and to maximize opportunities to create lasting change.

Freeman (2015) focused on five concepts that characterize an integrated landscape approach such as FLR, based on a structured literature review: multifunctionality, interdisciplinarity, participation, complexity and sustainability. In practice, integrated landscape approaches will involve trade-offs in order to achieve multifunctionality. Each of these concepts can be clearly defined in ways that are measurable, providing a way to operationalize them. New efforts need to focus on how to transform these concepts into indicators, criteria and standards that have relevance to local contexts. These indicators and standards can be used to monitor and evaluate landscape outcomes, as well as to guide implementation actions to achieve the multiple goals of FLR.

Abbas S, Nichol JE and Fischer GA. 2016. A 70-year perspective on tropical forest regeneration. Science of The Total Environment 544:544–52.

Adame M, Hermoso V, Perhans K, Lovelock C and Herrera‐Silveira J. 2015. Selecting cost‐effective areas for restoration of ecosystem services. Conservation Biology 29:493–502.

Ager AA, Vogler KC, Day MA and Bailey JD. 2017. Economic opportunities and trade-offs in collaborative forest landscape restoration. Ecological Economics 136:226–239.

Aldrich M and Sayer J. 2007. Landscape Outcomes Assessment Methodology “LOAM” – In Practice. WWF Forests for Life Program. IUCN, Gland, Switzerland. Accessed 13 October 2017. http://wwf.panda.org/?120980/Landscape-Outcome-Assessment-Methodology-LOAM-in-Practice

Armsworth PR. 2014. Inclusion of costs in conservation planning depends on limited datasets and hopeful assumptions. Annals of the New York Academy of Sciences 1322:61–76.

Arroyo-Rodriguez V, Melo F, Martinez-Ramos M, Bongers F, Chazdon R, Meave J, Norden N, Santos B, Leal I and Tabarelli M. 2017. Multiple successional pathways in human-modified tropical landscapes: New insights from forest succession, forest fragmentation and landscape ecology research. Biological Reviews 92:326–340.

Bagstad K, Villa F, Batker D, Harrison-Cox J, Voigt B and Johnson G. 2014. From theoretical to actual ecosystem services: Mapping beneficiaries and spatial flows in ecosystem service assessments. Ecology and Society 19(2):64.

Bagstad KJ, Semmens DJ, Waage S and Winthrop R. 2013. A comparative assessment of decision-support tools for ecosystem services quantification and valuation. Ecosystem Services 5:27–39.

Ball AA, Gouzerh A and Brancalion PH. 2014. Multi-scalar governance for restoring the Brazilian Atlantic Forest: A case study on small

landholdings in protected areas of sustainable development. Forests 5:599–619.

Ball IR, Possingham HP and Watts M. 2009. Forest rehabilitation in Indonesia: Where to after three decades? In: Moilanen A, Wilson KA and Possingham HP, eds. Spatial conservation prioritisation: Quantitative methods and computational tools. Oxford, UK: Oxford University Press. 185–95.

Banks-Leite C, Pardini R, Tambosi LR, Pearse WD, Bueno AA, Bruscagin RT, Condez TH, Dixo M, Igari AT and Martensen AC. 2014. Using ecological thresholds to evaluate the costs and benefits of set-asides in a biodiversity hotspot. Science 345:1041–45.

Baral H, Guariguata MR and Keenan RJ. 2016. A proposed framework for assessing ecosystem goods and services from planted forests. Ecosystem Services 22:260–268.

Baral H, Keenan RJ, Sharma SK, Stork NE and Kasel S. 2014. Spatial assessment and mapping of biodiversity and conservation priorities in a heavily modified and fragmented production landscape in north-central Victoria, Australia. Ecological Indicators 36:552–562.

Barlow J, Gardner TA, Araujo IS, Avila-Pires TC, Bonaldo AB, Costa JE, Esposito MC, Ferreira LV, Hawes J, Hernandez MIM, et al. 2007. Quantifying the biodiversity value of tropical primary, secondary and plantation forests. PNAS 104:18555–18560.

Barnett A, Fargione J and Smith MP. 2016. Mapping trade-offs in ecosystem services from reforestation in the Mississippi alluvial valley. BioScience 66:223–237.

Barr CM and Sayer JA. 2012. The political economy of reforestation and forest restoration in Asia–Pacific: Critical issues for REDD+. Biological Conservation 154:9–19.

Baynes J, Herbohn J and Dressler W. 2016. Power relationships: Their effect on the governance of community forestry in the Philippines. Land Use Policy 54:169–76.

References

http://wwf.panda.org/?120980/Landscape-Outcome-Assessment-Methodology-LOAM-in-Practice




Birch JC, Newton AC, Aquino CA, Cantarello E, Echeverría C, Kitzberger T, Schiappacasse I and Garavito NT. 2010. Cost-effectiveness of dryland forest restoration evaluated by spatial analysis of ecosystem services. PNAS 107:21925–21930.

Boedhihartono AK. 2012. Visualizing sustainable landscapes: Understanding and negotiating conservation and development trade-offs using visual techniques. Gland, Switzerland: IUCN.

Boumans R, Roman J, Altman I and Kaufman L. 2015. The Multiscale Integrated Model of Ecosystem Services (MIMES): Simulating the interactions of coupled human and natural systems. Ecosystem Services 12:30–41.

Brancalion PH and Chazdon RL. 2017. Beyond hectares: Four principles to guide reforestation in the context of tropical forest and landscape restoration. Restoration Ecology 25:491–96.

Brancalion PHS and van Melis J. 2017. On the need for innovation in ecological restoration. Annals of the Missouri Botanical Garden 102:227–36.

Breckheimer I, Haddad NM, Morris WF, Trainor AM, Fields WR, Jobe R, Hudgens BR, Moody A and Walters JR. 2014. Defining and evaluating the umbrella species concept for conserving and restoring landscape connectivity. Conservation Biology 28:1584–1593.

Bryan BA, Perry L and Gerner D. 2004. An interactive spatial optimisation tool for systematic landscape restoration. International Congress on Environmental Modelling and Software. 47. https://scholarsarchive.byu.edu/iemssconference/2004/all/47 Accessed 22 May, 2017.

Buckingham K, Ray S, Arakwiye B, Morales AG, Singh R, Maneerattana D, Wicaksono S and Chrysolite H. 2018. Mapping social landscapes: A guide to restoration opportunities mapping. Washington, DC: World Resources Institute.

Buckingham K, Ray S, Stolle F and Zoveda F. 2017. Measuring progress for forest and landscape restoration. Working Paper Version 1.0. Washington, DC: World Resources Institute and the Food and Agriculture Organization of the United Nations.

Budiharta S, Meijaard E, Wells JA, Abram NK and Wilson KA. 2016. Enhancing feasibility: Incorporating a socio-ecological systems framework into restoration planning. Environmental Science and Policy 64:83–92.

Bullock JM and Ding H. 2018. A guide to selecting ecosystem service models for decision-making:

Lessons from sub-Saharan Africa. Washington, DC: World Resources Institute.

Bürgi M, Ali P, Chowdhury A, Heinimann A, Hett C, Kienast F, Mondal MK, Upreti BR and Verburg PH. 2017. Integrated Landscape Approach: Closing the gap between theory and application. Sustainability 9:1371.

Carwardine J, Hawkins C, Polglase P, Possingham HP, Reeson A, Renwick AR, Watts M and Martin TG. 2015. Spatial priorities for restoring biodiverse carbon forests. BioScience 65:372–82.

Cerbu GA, Swallow BM and Thompson DY. 2011. Locating REDD: A global survey and analysis of REDD readiness and demonstration activities. Environmental Science & Policy 14:168–80.

Chazdon RL. 2017. Landscape restoration, natural regeneration, and the forests of the future. Annals of the Missouri Botanical Garden 102:251–57.

Chazdon RL. 2014. Second growth: The promise of tropical forest regeneration in an age of deforestation. Chicago: University of Chicago Press.

Chazdon RL. 2013. Making tropical succession and landscape reforestation successful. Journal of Sustainable Forestry 32:649–58.

Chazdon RL and Guariguata MR. 2016. Natural regeneration as a tool for large-scale forest restoration in the tropics: Prospects and challenges. Biotropica 48:716–30.

Chazdon RL, Brancalion PH, Lamb D, Laestadius L, Calmon M and Kumar C. 2017. A policy‐driven knowledge agenda for global forest and landscape restoration. Conservation Letters 10:125–32.

Chazdon RL, Brancalion PH, Laestadius L, Bennett-Curry A, Buckingham K, Kumar C, Moll-Rocek J, Vieira ICG and Wilson SJ. 2016. When is a forest a forest? Forest concepts and definitions in the era of forest and landscape restoration. Ambio 45:538–50.

Chokkalingam U, Carandang AP, Pulhin JM, Lasco RD, Peras RJJ and Toma T. 2006. One century of forest rehabilitation in the Philippines: Approaches, outcomes and lessons. Bogor, Indonesia: CIFOR.

Christin ZL, Bagstad KJ and Verdone MA. 2016. A decision framework for identifying models to estimate forest ecosystem services gains from restoration. Forest Ecosystems 3:1–12.

Ciecko L, Kimmett D, Saunders J, Katz R, Wolf KL, Bazinet O, Richardson J, Brinkley

https://scholarsarchive.byu.edu/iemssconference/2004/all/47 Accessed 22 May, 2017



https://books.google.com.pe/books?hl=en&lr=&id=EYhFAwAAQBAJ&oi=fnd&pg=PR7&dq=chazdon+second+growth&ots=7OKDtvBzl6&sig=F-YcZRcaKfgsQVNAh1F4hJf3a4s




W and Blahna DJ. 2016. Forest Landscape Assessment Tool (FLAT): Rapid assessment for land management. General technical report. PNW-GTR-941. Portland, Oregon: US Department of Agriculture, Forest Service, Pacific Northwest Research Station.

Costanza R, D’arge R, De Groot R, Farber S, Grasso M, Hannon B, Limburg K, Naeem S, O’Neill RV and Paruelo J. 1997. The value of the world’s ecosystem services and natural capital. Nature 387:253–260.

Crossman ND and Bryan BA. 2009. Identifying cost-effective hotspots for restoring natural capital and enhancing landscape multifunctionality. Ecological Economics 68:654–68.

Crossman ND and Bryan BA. 2006. Systematic landscape restoration using integer programming. Biological Conservation 128:369–83.

Crouzeilles R, Beyer HL, Mills M, Grelle CE and Possingham HP. 2015. Incorporating habitat availability into systematic planning for restoration: A species‐specific approach for Atlantic forest mammals. Diversity and Distributions 21:1027–37.

Daily GC. 1997. Nature’s services: Societal dependence on natural ecosystems. Washington, DC: Island Press.

De Ridder W, Turnpenny J, Nilsson M and Von Raggamby A. 2007. A framework for tool selection and use in integrated assessment for sustainable development. Journal of Environmental Assessment Policy and Management 9:423–41.

Derak M, Cortina J, Taiqui L and Aledo A. 2017. A proposed framework for participatory forest restoration in semiarid areas of North Africa. Restoration Ecology. DOI: 10.1111/rec.12486

Ding H, Faruqi S, Wu A, Altamirano JC, Anchondo Ortega A, Verdone M, Zamora Cristales R, Chazdon R and Vergara W. 2017. Roots of prosperity: The economics and finance of restoring land. Washington, DC: World Resources Institute.

Dirzo R, Young HS, Galetti M, Ceballos G, Isaac NJ and Collen B. 2014. Defaunation in the Anthropocene. Science 345:401–6.

Douthwaite B, Alvarez, S, Cook, S, Davies, R, George, P, Howell, J, Mackay, R and Rubiano, J. 2007. Participatory impact pathways analysis: a practical application of program theory in research-for-development. The Canadian Journal of Program Evaluation 22:127.

Duarte GT, Ribeiro MC and Paglia AP. 2016. Ecosystem services modeling as a tool for defining priority areas for conservation. PLoS One 11:e0154573.

Evans K and Guariguata MR. 2016. Success from the ground up: Participatory monitoring and forest restoration. Occasional Paper 169. Bogor, Indonesia: CIFOR.

[FAO] Food and Agriculture Organization of the United Nations. 2016. Free, prior, informed consent: An indigenous peoples’ right and a good practice for local communities. Practitioners manual. Rome: FAO.

[FAO] Food and Agriculture Organization of the United Nations. 2010. Country report for Malawi. Rome: FAO. Accessed 3 May, 2017. http://www.fao.org/docrep/013/al557E/al557E.pdf.

[FAO and UNCCD] Food and Agriculture Organization of the United Nations and Global Mechanism of the United Nations Convention to Combat Desertification. 2015. Sustainable financing for forest and landscape restoration: Opportunities, challenges and the way forward. Discussion paper. Rome: FAO and UNCCD.

Farber S, Costanza R, Childers DL, Erickson J, Gross K, Grove M, Hopkinson CS, Kahn J, Pincetl S, Troy A, Warren P and Wilson M. 2016. Linking ecology and economics for ecosystem management. BioScience 56:121–133.

Faruqi S and Landsberg F. 2017. Attracting private investment to landscape restoration: A roadmap. Washington, DC: World Resources Institute. Accessed 4 April 2017. http://www.wri.org/publication/attracting-private-investment-to-landscape-restoration

Fernández IC and Morales NS. 2016. A spatial multicriteria decision analysis for selecting priority sites for plant species restoration: A case study from the Chilean biodiversity hotspot. Restoration Ecology 24:599–608.

Freeman OE, Duguma LA and Minang PA. 2015. Operationalizing the integrated landscape approach in practice. Ecology and Society 20:24ff.

Garcia C, Dray A and Waeber P. 2016. Learning begins when the game is over: using games to embrace complexity in natural resources management. GAIA-Ecological Perspectives for Science and Society 25:289–91.

García-Barrios L, García-Barrios R, Waterman A and Cruz-Morales J. 2011. Social

http://www.fao.org/docrep/013/al557E/al557E.pdf

http://www.fao.org/docrep/013/al557E/al557E.pdf




dilemmas and individual/group coordination strategies in a complex rural land-use game. International Journal of the Commons 5:364–87.

Gardner TA, Barlow J, Chazdon R, Ewers RM, Harvey CA, Peres CA and Sodhi NS. 2009. Prospects for tropical forest biodiversity in a human modified world. Ecology Letters 12:561–582.

Ghazoul J and Chazdon RL. 2017. Degradation and recovery in changing forest landscapes: A multi-scale conceptual framework. Annual Review of Environment and Resources 42:161–88.

Giannini TC, Giulietti AM, Harley RM, Viana PL, Jaffe R, Alves R, Pinto CE, Mota NF, Caldeira CF and Imperatriz‐Fonseca VL. 2017. Selecting plant species for practical restoration of degraded lands using a multiple‐trait approach. Austral Ecology 42:510–21.

Gimona A and Van Der Horst D. 2007. Mapping hotspots of multiple landscape functions: A case study on farmland afforestation in Scotland. Landscape Ecology 22:1255–64.

Gourevitch JD, Hawthorne PL, Keeler BL, Beatty CR, Greve M and Verdone MA. 2016. Optimizing investments in national-scale forest landscape restoration in Uganda to maximize multiple benefits. Environmental Research Letters 11:114027.

Gregorio N, Herbohn J, Harrison S, Pasa A and Ferraren A. 2017. Regulating the quality of seedlings for forest restoration: Lessons from the National Greening Program in the Philippines. Small-scale Forestry 16:83–102.

Greve M, Reyers B, Lykke AM and Svenning J-C. 2013. Spatial optimization of carbon-stocking projects across Africa integrating stocking potential with co-benefits and feasibility. Nature Communications 4: 2975

Hajdu F and Fischer K. 2017. Problems, causes and solutions in the forest carbon discourse: A framework for analysing degradation narratives. Climate and Development 9:537–47.

Hajdu F, Penje O and Fischer K. 2016. Questioning the use of ‘degradation’ in climate mitigation: A case study of a forest carbon CDM project in Uganda. Land Use Policy 59:412–22.

Han Z, Song W, Deng X and Xu X. 2017. Trade-offs and synergies in ecosystem service within the Three-Rivers headwater region, China. Water 9:588.

Holmgren M and Scheffer M. 2017. To tree or not to tree: Cultural views from ancient Romans to modern ecologists. Ecosystems 20:62–8

Iftekhar MS, Polyakov M, Ansell D, Gibson F and Kay GM. 2017. How economics can further the success of ecological restoration. Conservation Biology 31:261–268.

Ikin K, Tulloch A, Gibbons P, Ansell D, Seddon J and Lindenmayer D. 2016. Evaluating complementary networks of restoration plantings for landscape-scale occurrence of temporally dynamic species. Conservation Biology 30:1027–37.

Instituto Escolhas. 2016. Quanto é? Plantar Floresta. São Paulo, Brazil: Instituto Escolhas. Accessed 7 October 2017. http://quantoefloresta.escolhas.org

[IPBES] Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. 2017. Update on the classification of nature’s contributions to people. Plenary of the Intergovernmental Science-Policy. Platform on Biodiversity and Ecosystem Services 7–10 March 2017, Bonn, Germany. https://www.ipbes.net/sites/default/files/downloads/pdf/ipbes-5-inf-24.pdf

IUCN] International Union for Conservation of Nature. 2015. Bonn Challenge Commitments. Washington, DC: IUCN. Accessed 4 October 2017. http://www.bonnchallenge.org/commitments

[IUCN and WRI] International Union for Conservation of Nature and World Resources Institute. 2014. A guide to the Restoration Opportunities Assessment Methodology (ROAM): Assessing forest landscape restoration opportunities at the national or sub-national level. Working paper (road-test edition). Gland, Switzerland: IUCN.

Kang M and Kim S. 2015. Development of North Korea’s reforestation strategies through benchmarking six Asian countries’ cases. Forest Science and Technology 11:104–10.

Keenleyside K, Dudley N, Cairns S, Hall C and Stolton S. 2012. Ecological restoration for protected areas: Principles, guidelines and best practices. Gland, Switzerland: IUCN.

Kennedy CM, Miteva DA, Baumgarten L, Hawthorne PL, Sochi K, Polasky S, Oakleaf JR, Uhlhorn EM and Kiesecker J. 2016. Bigger is better: Improved nature conservation and economic returns from landscape-level mitigation. Science Advances 2:e1501021.

Kindt R, Graudal L, Jamnadass R, Lillesø J-PB, Orwa C and van Breugel P. 2015a. Useful tree species for Eastern Africa: A species selection tool based on the VECEA map. Version 2.0.

http://quantoefloresta.escolhas.org

http://quantoefloresta.escolhas.org

https://www.ipbes.net/sites/default/files/downloads/pdf/ipbes-5-inf-24.pdf



http://www.bonnchallenge.org/commitments

http://www.bonnchallenge.org/commitments


Bogor, Indonesia: World Agroforestry Centre. https://www.worldagroforestry.org/publication/useful-tree-species-eastern-africa-species-selection-tool-based-vecea-map-version-20

Kindt R, Ordoñez J, Smith E, Orwa C, Mosoti B, John I, Chege J, Harja D, Kehlenbeck K, Luedeling E, et al. 2015b. ICRAF Species Switchboard Version 1.2. Bogor, Indonesia: World Agroforestry Centre. http://www.forskningsdatabasen.dk/en/catalog/2304047265

Kukkala AS and Moilanen A. 2013. Core concepts of spatial prioritisation in systematic conservation planning. Biological Reviews 88:443–64.

Laestadius L, Buckingham K, Maginnis S and Saint-Laurent C. 2015. Back to Bonn and beyond: A history of forest landscape restoration and an outlook for the future. Unasylva 245 66:11–8.

Laestadius L, Maginnis S, Minnemeyer S, Potapov P, Saint-Laurent C and Sizer N. 2012. Mapping opportunities for forest landscape restoration. Unasylva 62:47–8.

Lagabrielle E, Rouget M, Le Bourgeois T, Payet K, Durieux L, Baret S, Dupont J and Strasberg D. 2011. Integrating conservation, restoration and land-use planning in islands: An illustrative case study in Réunion Island (Western Indian Ocean). Landscape and Urban Planning 101:120–30.

Latawiec AE, Crouzeilles R, Brancalion PHS, Rodrigues RR, Sansevero JBB, Dos Santos JS, Mills M, Nave AG and Strassburg BBN. 2016. Natural regeneration and biodiversity: A global meta-analysis and implications for spatial planning. Biotropica 48:844–55.

Le HD, Smith C and Herbohn J. 2014. What drives the success of reforestation projects in tropical developing countries? The case of the Philippines. Global Environmental Change 24:334–48.

Le HD, Smith C, Herbohn J and Harrison S. 2012. More than just trees: Assessing reforestation success in tropical developing countries. Journal of Rural Studies 28:5–19.

Lele S and Srinivasan V. 2013. Disaggregated economic impact analysis incorporating ecological and social trade-offs and techno-institutional context: A case from the Western Ghats of India. Ecological Economics 91:98–112.

Locatelli B, Imbach P and Wunder S. 2014. Synergies and trade-offs between ecosystem services in Costa Rica. Environmental Conservation 41:27–36.

Lynam T, De Jong W, Sheil D, Kusumanto T and Evans K. 2007. A review of tools for

incorporating community knowledge, preferences, and values into decision making in natural resources management. Ecology and Society 12:5.

Macmillan DC, Harley D and Morrison R. 1998. Cost-effectiveness analysis of woodland ecosystem restoration. Ecological Economics 27:313–324.

Maginnis S and Jackson W. 2005. Balancing restoration and development. ITTO Tropical Forest Update 15:4–6.

Malawi Ministry of Natural Resources, Energy and Mining. 2017. Forest Landscape Restoration Opportunities Assessment for Malawi. NFLRA (Malawi), IUCN, WRI.

Mansourian S and Vallauri D. 2014. Restoring forest landscapes: Important lessons learnt. Environmental Management 53:241–51.

Mansourian S, Stanturf JA, Derkyi MaA and Engel VL. 2017. Forest landscape restoration: Increasing the positive impacts of forest restoration or simply the area under tree cover? Restoration Ecology doi: 10.1111/rec.12489

Mansourian S, Vallauri D and Dudley N. 2005. Forest restoration in landscapes: Beyond planting trees. New York: Springer.

Margules CR and Pressey RL. 2000. Systematic conservation planning. Nature 405:243.

Martinez-Harms MJ, Bryan BA, Balvanera P, Law EA, Rhodes JR, Possingham HP and Wilson KA. 2015. Making decisions for managing ecosystem services. Biological Conservation 184:229–238.

McBride MF, Wilson KA, Burger J, Fang Y-C, Lulow M, Olson D, O’Connell M and Possingham HP. 2010. Mathematical problem definition for ecological restoration planning. Ecological Modelling 221:2243–50.

Mccall MK. 2016. Beyond ‘landscape’ in REDD+: The imperative for ‘territory’. World Development 85:58–72.

McDonald T, Gann G, Jonson J and Dixon K. 2016. International standards for the practice of ecological restoration–including principles and key concepts. Washington, DC: Society for Ecological Restoration.

McRae BH, Hall SA, Beier P and Theobald DM. 2012. Where to restore ecological connectivity? Detecting barriers and quantifying restoration benefits. PLoS One 7:e52604.

MEA. 2005. Ecosystems and Human Well-being: Synthesis. Washington, DC: Island Press.

Meli P, Martínez‐Ramos M, Rey‐Benayas JM and Carabias J. 2014. Combining ecological,

https://www.worldagroforestry.org/publication/useful-tree-species-eastern-africa-species-selection-tool-based-vecea-map-version-20



http://www.forskningsdatabasen.dk/en/catalog/2304047265

http://www.forskningsdatabasen.dk/en/catalog/2304047265


social and technical criteria to select species for forest restoration. Applied Vegetation Science 17:744–53.

Mendizabal, E. 2010. The alignment, interest and influence matrix (AIIM) guidance note. ODI Manual/Toolkit. London: ODI.

Metzger JP, Esler K, Krug C, Arias M, Tambosi L, Crouzeilles R, Acosta AL, Brancalion PH, D’Albertas F and Duarte GT. 2017. Best practice for the use of scenarios for restoration planning. Current Opinion in Environmental Sustainability 29:14–25.

Moilanen A and Kujala H. 2008. Zonation spatial conservation planning framework and software v. 2.0: user manual. Helsinki, Finland: Edita. Accessed 4 May, 2017. http://www.helsinki.fi/science/metapop/softward.

Monroe AS and Butler WH. 2016. Responding to a policy mandate to collaborate: Structuring collaboration in the collaborative forest landscape restoration program. Journal of Environmental Planning and Management 59:1054–72.

Moreno-Mateos D, Barbier EB, Jones PC, Jones HP, Aronson J, López-López JA, Mccrackin ML, Meli P, Montoya D and Rey Benayas JM. 2017. Anthropogenic ecosystem disturbance and the recovery debt. Nature Communications 8:14163.

Mulligan M, Guerry A, Arkema K, Bagstad K and Villa F. 2010. Capturing and quantifying the flow of ecosystem services. In: Silvestri S and Kershaw F, eds. Framing the Flow: Innovative Approaches to Understand, Protect and Value Ecosystem Services Across Linked Habitats. Cambridge, UK: UNEP World Conservation Monitoring Centre.

Murcia C, Guariguata MR, Andrade Á, Andrade GI, Aronson J, Escobar EM, Etter A, Moreno FH, Ramírez W and Montes E. 2016. Challenges and prospects for scaling‐up ecological restoration to meet international commitments: Colombia as a case study. Conservation Letters 9:213–20.

Naidoo R and Ricketts TH. 2006. Mapping the economic costs and benefits of conservation. PLoS Biology 4:2153–2164.

Nawir AA, Murniati and Rumboko L. 2007. Forest rehabilitation in Indonesia: Where to after three decades? Bogor, Indonesia: CIFOR.

Nelson E, Mendoza G, Regetz J, Polasky S, Tallis H, Cameron DR, Chan KMA, Daily GC, Goldstein J and Kareiva PM. 2009. Modeling multiple ecosystem services, biodiversity conservation, commodity production, and

tradeoffs at landscape scales. Frontiers in Ecology and the Environment 7:4–11.

Nelson E, Polasky S, Lewis DJ, Plantinga AJ, Lonsdorf E, White D, Bael D and Lawler JJ. 2008. Efficiency of incentives to jointly increase carbon sequestration and species conservation on a landscape. PNAS 105(28):9471–9476.

Nunes FS, Soares-Filho BS, Rajão R and Merry F. 2017. Enabling large-scale forest restoration in Minas Gerais state, Brazil. Environmental Research Letters 12:044022

Orsi F and Geneletti D. 2010. Identifying priority areas for forest landscape restoration in Chiapas (Mexico): An operational approach combining ecological and socioeconomic criteria. Landscape and Urban Planning 94:20–30.

Orsi F, Geneletti D and Newton AC. 2011. Towards a common set of criteria and indicators to identify restoration priorities: An expert panel-based approach. Ecological Indicators 11:337–47.

Oteros-Rozas E, Martín-López B, Daw T, Bohensky EL, Butler J, Hill R, Martin-Ortega J, Quinlan A, Ravera F, Ruiz-Mallén I, Thyresson M, Mistry J, Palomo I, Peterson GD, Plieninger T, Waylen KA, Beach D, Bohnet IC, Hamann M, Hanspach J, Hubacek K, Lavorel S and Vilardy S. 2015. Participatory scenario planning in place-based social-ecological research: insights and experiences from 23 case studies. Ecology and Society 20(4):32. http://dx.doi.org/10.5751/ES-07985-200432

Palomo I, Felipe-Lucia MR, Bennett EM, Martín-López B and Pascual U. 2016. Disentangling the pathways and effects of ecosystem service co-production. Advances in Ecological Research 54:245–283.

Pardini R, De Arruda Bueno A, Gardner TA, Prado PI and Metzger JP. 2010. Beyond the fragmentation threshold hypothesis: regime shifts in biodiversity across fragmented landscapes. PloS One 5:e13666.

Pascual U, Balvanera P, Díaz S, Pataki G, Roth E, Stenseke M, Watson RT, Dessane EB, Islar M and Kelemen E. 2017a. Valuing nature’s contributions to people: The IPBES approach. Current Opinion in Environmental Sustainability 26:7–16.

Pascual U, Palomo I, Adams W, Chan K, Daw T, Garmendia E, Gómez-Baggethun E, De Groot R, Mace G and Martin-Lopez B. 2017b.

http://www.helsinki.fi/science/metapop/softward

http://www.helsinki.fi/science/metapop/softward

http://dx.doi.org/10.5751/ES-07985-200432

http://dx.doi.org/10.5751/ES-07985-200432


Off-stage ecosystem service burdens: A blind spot for global sustainability. Environmental Research Letters 12:7.

Paudyal K, Baral H and Keenan RJ. 2016. Local actions for the common good: can the application of the ecosystem services concept generate improved societal outcomes from natural resource management? Land Use Policy 56:327–32.

Peh KS-H, Balmford A, Bradbury RB, Brown C, Butchart SH, Hughes FM, Stattersfield A, Thomas DH, Walpole M and Bayliss J. 2013. TESSA: A toolkit for rapid assessment of ecosystem services at sites of biodiversity conservation importance. Ecosystem Services 5:51–57.

Pistorius T, Carodenuto S and Wathum G. 2017. Implementing forest landscape restoration in Ethiopia. Forests 8:61.

Rappaport DI, Tambosi LR and Metzger JP. 2015. A landscape triage approach: Combining spatial and temporal dynamics to prioritize restoration and conservation. Journal of Applied Ecology 52:590–601.

Redhead JW, May L, Oliver TH, Hamel P, Sharp R and Bullock JM. 2018. National scale evaluation of the InVEST nutrient retention model in the United Kingdom. Science of The Total Environment 610:666–677.

Reed J, Van Vianen J, Barlow J and Sunderland T. 2017. Have integrated landscape approaches reconciled societal and environmental issues in the tropics? Land Use Policy 63:481–92.

Reed J, Van Vianen J, Deakin EL, Barlow J and Sunderland T. 2016. Integrated landscape approaches to managing social and environmental issues in the tropics: Learning from the past to guide the future. Global Change Biology 22:2540–54.

Reed MS. 2008. Stakeholder participation for environmental management: A literature review. Biological Conservation 141:2417–31.

Reid JL, Wilson SJ, Bloomfield GS, Cattau ME, Fagan ME, Holl KD and Zahawi RA. 2017. How long do restored ecosystems persist? Annals of the Missouri Botanical Garden 102:258–65.

Renwick AR, Robinson CJ, Martin TG, May T, Polglase P, Possingham HP and Carwardine J. 2014. Biodiverse Planting for Carbon and Biodiversity on Indigenous Land. PloS One 9: e91281.

Republic of Malawi, Ministry of Natural Resources, Energy and Mining. 2017. National Forest Landscape Restoration Strategy, 44.

Reubens B, Moeremans C, Poesen J, Nyssen J, Tewoldeberhan S, Franzel S, Deckers J, Orwa C and Muys B. 2011. Tree species selection for land rehabilitation in Ethiopia: from fragmented knowledge to an integrated multi-criteria decision approach. Agroforestry Systems 82:303–30.

Reyers B, Biggs R, Cumming GS, Elmqvist T, Hejnowicz AP and Polasky S. 2013. Getting the measure of ecosystem services: A social-ecological approach. Frontiers in Ecology and the Environment 11:268–273.

Richards RC, Rerolle J, Aronson J, Pereira PH, Gonçalves H and Brancalion PH. 2015. Governing a pioneer program on payment for watershed services: Stakeholder involvement, legal frameworks and early lessons from the Atlantic forest of Brazil. Ecosystem Services 16:23–32.

Ruckelshaus M, Mckenzie E, Tallis H, Guerry A, Daily G, Kareiva P, Polasky S, Ricketts T, Bhagabati N and Wood SA. 2015. Notes from the field: Lessons learned from using ecosystem service approaches to inform real-world decisions. Ecological Economics 115:11–21.

Rwanda Ministry of Resources. 2014. Forest landscape restoration opportunity assessment for Rwanda. Nairobi, Kenya: Minirena, IUCN, WRI.

Sabogal C, Besacier C and Mcguire D. 2015. Forest and landscape restoration: Concepts, approaches and challenges for implementation. Unasylva 245 66:3–10.

Salata S, Garnero G, Barbieri CA, and Giaimo C. 2017. The integration of ecosystem services in planning: An evaluation of the nutrient retention model using InVEST software. Land 6:48.

Sayer JA, Margules C, Boedhihartono AK, Sunderland T, Langston JD, Reed J, Riggs R, Buck LE, Campbell BM, Kusters K and Elliott C. 2017. Measuring the effectiveness of landscape approaches to conservation and development. Sustainability Science 12:465–76.

Sayer J, Sunderland T, Ghazoul J, Pfund J-L, Sheil D, Meijaard E, Venter M, Boedhihartono AK, Day M and Garcia C. 2013. Ten principles for a landscape approach to reconciling agriculture, conservation, and other competing land uses. Proceedings of the National Academy of Sciences 110:8349–56.

Scales BR and Marsden SJ. 2008. Biodiversity in small-scale tropical agroforests: A review of species richness and abundance shifts and the factors influencing them. Environmental Conservation 35:160–172.


Schulz JJ and Schröder B. 2017. Identifying suitable multifunctional restoration areas for forest landscape restoration in Central Chile. Ecosphere 8:e01644.

Sherrouse BC, Semmens DJ and Clement JM. 2014. An application of Social Values for Ecosystem Services (SolVES) to three national forests in Colorado and Wyoming. Ecological Indicators 36:68–79.

Silva D and Nunes S. 2017. Avaliação e modelagem econômica da restauração florestal no estado do Pará. Belém, Brasil: Imazon.

Simula, M. 2008. Financing flows and needs to implement the non-legally binding instrument on all types of forests. UNFF/World Bank side event during UNFCCCC COP 14. Poznan, Poland: PROFOR/World Bank.

Song C-H, Lee W-K, Choi H-A, Jeon S-W, Kim J-U, Kim J-S and Kim J-T. 2015. Application of InVEST water yield model for assessing forest water provisioning ecosystem service. Journal of the Korean Association of Geographic Information Studies 18:120–134.

Stanturf J, Mansourian S and Kleine M. 2017. Implementing forest landscape restoration, a practitioner’s guide. Vienna, Austria: International Union of Forest Research Organizations (IUFRO).

Stanturf JA, Kant P, Lillesø J-PB, Mansourian S, Kleine M, Graudal L and Madsen P. 2015. Forest landscape restoration as a key component of climate change mitigation and adaptation. IUFRO World Series Volume 34. Vienna, Austria: International Union of Forest Research Organizations (IUFRO). 72.

Stefanes M, Ochoa-Quintero JM, Roque F, Sugai LS, Tambosi L, Lourival R and Laurance S. 2016. Incorporating resilience and cost in ecological restoration strategies at landscape scale. Ecology and Society 21:54.

Strassburg BN, Barros FSM, Couzeilles R, Iribarrem A, Santos JS, Silva D, Sansevero JBB, Alves-Pinto H, Feltran-Barbieri R and Latawiec A. 2016. The role of natural regeneration to ecosystem services provision and habitat availability: A case study in the Brazilian Atlantic Forest. Biotropica 48:890–899.

Sukhdev P. 2008. The economics of ecosystems and biodiversity: Interim report. Cambridge, UK: European Community.

Sutherland IJ, Bennett EM and Gergel SE. 2016. Recovery trends for multiple ecosystem services reveal non-linear responses and long-term

tradeoffs from temperate forest harvesting. Forest Ecology and Management 374:61–70.

Talema A, Poesen J, Muys B, Reubens B, Dibaba H and Diels J. 2017. Multi‐criteria‐based plant species selection for gully and riverbank stabilization in a sub‐humid tropical area. Land Degradation & Development 28:1675–86.

Tallis H, Mooney H, Andelman S, Balvanera P, Cramer W, Karp D, Polasky S, Reyers B, Ricketts T and Running S. 2012. A global system for monitoring ecosystem service change. BioScience 62:977–986.

Tallis H and Polasky S. 2009. Mapping and valuing ecosystem services as an approach for conservation and natural‐resource management. Annals of the New York Academy of Sciences 1162:265–283.

Tambosi LR, Martensen AC, Ribeiro MC and Metzger JP. 2014. A framework to optimize biodiversity restoration efforts based on habitat amount and landscape connectivity. Restoration Ecology 22:169–177.

Thomas E. 2016. A tool to guide species and seed selection for the restoration of seasonally dry tropical forest in Colombia. Rome: Bioversity. Accessed 12 October 2017. https://www.bioversityinternational.org/news/detail/a-tool-to-guide-species-and-seed-selection-for-the-restoration-of-seasonally-dry-tropical-forest-in-colombia/

Thomson JR, Moilanen AJ, Vesk PA, Bennett AF and Mac Nally R. 2009. Where and when to revegetate: A quantitative method for scheduling landscape reconstruction. Ecological Applications 19:817–28.

Tobón W, Urquiza‐Haas T, Koleff P, Schröter M, Ortega‐Álvarez R, Campo J, Lindig Cisneros R, Sarukhán J and Bonn A. 2017. Restoration planning to guide Aichi targets in a megadiverse country. Conservation Biology 31: 1086–97

Trabucchi M, O’Farrell PJ, Notivol E and Comín FA. 2014. Mapping ecological processes and ecosystem services for prioritizing restoration efforts in a semi-arid Mediterranean river basin. Environmental Management 53:1132–45.

Turner KG, Anderson S, Gonzales-Chang M, Costanza R, Courville S, Dalgaard T, Dominati E, Kubiszewski I, Ogilvy S and Porfirio L. 2016. A review of methods, data, and models to assess changes in the value of ecosystem services from land degradation and restoration. Ecological Modelling 319:190–207.

Urgenson LS, Ryan CM, Halpern CB, Bakker JD, Belote RT, Franklin JF, Haugo RD, Nelson CR

https://www.bioversityinternational.org/news/detail/a-tool-to-guide-species-and-seed-selection-for-the-restoration-of-seasonally-dry-tropical-forest-in-colombia/






and Waltz AE. 2017. Visions of restoration in fire-adapted forest landscapes: Lessons from the collaborative forest landscape restoration program. Environmental Management 59:338–53.

Uribe D, Geneletti D, Castillo RFD and Orsi F. 2014. Integrating stakeholder preferences and GIS-based multicriteria analysis to identify forest restoration priorities. Sustainability 6:935–51.

Vallauri D, Aronson J and Dudley N. 2005. An attempt to develop a framework for restoration planning. In: Mansourian S, Vallauri D and Dudley N, eds. Forest restoration in landscapes: Beyond planting trees. Springer, New York: Springer. 65–70.

Van Oosten C. 2013. Restoring landscapes—governing place: A learning approach to forest landscape restoration. Journal of Sustainable Forestry 32:659–76.

Van Oosten C, Gunarso P, Koesoetjahjo I and Wiersum F. 2014. Governing forest landscape restoration: Cases from Indonesia. Forests 5:1143-62.

Veldman JW, Overbeck GE, Negreiros D, Mahy G, Le Stradic S, Fernandes GW, Durigan G, Buisson E, Putz FE and Bond WJ. 2015. Where tree planting and forest expansion are bad for biodiversity and ecosystem services. BioScience 65:1011–18.

Verdone M and Seidl A. 2017. Time, space, place, and the Bonn Challenge global forest restoration target. Restoration Ecology 25: 903–11.

Villa F, Bagstad KJ, Voight B, Johnson GW, Portela R, Honzák M and Batker D. 2014. A methodology for adaptable and robust ecosystem services assessment. PLoS One 9:e91001. doi:91010.91371/journal.pone.0091001

Vogler KC, Ager AA, Day MA, Jennings M and Bailey JD. 2015. Prioritization of forest restoration projects: Tradeoffs between wildfire protection, ecological restoration and economic objectives. Forests 6:4403–20.

Wada CA, Bremer LL, Burnett K, Trauernicht C, Giambelluca T, Mandle L, Parsons E, Weil C, Kurashima N and Ticktin T. 2017. Estimating cost-effectiveness of Hawaiian dry forest restoration using spatial changes in water yield and landscape flammability under climate change. Pacific Science 71:401–424.

Weiss CH. 1997. How can theory-based evaluation make greater headway? Evaluation Review 21:501–24.

Wendland KJ, Honzák M, Portela R, Vitale B, Rubinoff S and Randrianarisoa J. 2010.

Targeting and implementing payments for ecosystem services: Opportunities for bundling biodiversity conservation with carbon and water services in Madagascar. Ecological Economics 69:2093–2107.

Willemen L, Kozar R, Desalegn A and Buck LE. 2014. Spatial planning and monitoring of landscape interventions: Maps to link people with their landscapes. A User’s Guide. Washington, DC: EcoAgriculture Partners.

Wilson KA, Lulow M, Burger J, Fang YC, Andersen C, Olson D, O’Connell M and Mcbride MF. 2011. Optimal restoration: Accounting for space, time and uncertainty. Journal of Applied Ecology 48:715–25.

van der Wolf J, Gram G, Bukomeko H, Mukasa D, Giller O, Kriabo E, Angebault C, Vaast P, Asare R and Jassogne L. 2017. The shade tree advice tool. CCAFS info note. Copenhagen, Denmark: CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). https://cgspace.cgiar.org/handle/10568/80473

Woollen E, Ryan CM, Baumert S, Vollmer F, Grundy I, Fisher J, Fernando J, Luz A, Ribeiro N and Lisboa SN. 2016. Charcoal production in the Mopane woodlands of Mozambique: What are the trade-offs with other ecosystem services? Philosophical Transactions of the Royal Society B 371:20150315.

[WRI] World Resources Institute. 2015. The restoration diagnostic: A method for developing forest landscape restoration strategies by rapidly assessing the status of key success factors. Washington, DC: WRI.

[WWF and IUCN] World Wide Fund for Nature and International Union for Conservation of Nature. 2000. Minutes of the Forests Reborn international workshop on forest restoration. Segovia, Spain, 3-5 July 2000. Gland, Switzerland: IUCN. Accessed 15 April 2017. https://cmsdata.iucn.org/downloads/flr_segovia.pdf

Yoshioka A, Akasaka M and Kadoya T. 2014. Spatial prioritization for biodiversity restoration: A simple framework referencing past species distributions. Restoration Ecology 22:185–195.

Zhou P, Luukkanen O, Tokola T and Nieminen J. 2008. Vegetation dynamics and forest landscape restoration in the Upper Min River watershed, Sichuan, China. Restoration Ecology 16:348–58.

Zweiner VP, Padial AA, Marques MCM, Faleiro FV, Loyola R and Peterson AT. 2017. Planning for conservation and restoration under climate and land use change in the Brazilian Atlantic Forest. Diversity and Distributions 27:1–12.

https://cgspace.cgiar.org/handle/10568/80473

https://cgspace.cgiar.org/handle/10568/80473

https://cmsdata.iucn.org/downloads/flr_segovia.pdf

https://cmsdata.iucn.org/downloads/flr_segovia.pdf

The CGIAR Research Program on Forests, Trees and Agroforestry (FTA) is the world’s largest research for development program to enhance the role of forests, trees and agroforestry in sustainable development and food security and to address climate change. CIFOR leads FTA in partnership with Bioversity International, CATIE, CIRAD, ICRAF, INBAR and TBI.

cifor.org forestsnews.cifor.org

Decision-making bodies at all scales face an urgent need to conserve remaining forests, and reestablish forest cover in deforested and degraded forest landscapes. The scale of the need, and the opportunity to make a difference, is enormous. Degradation is often viewed as ‘the problem’, and restoration as ‘the solution’. But, rather than being a goal, restoration is the means to achieve many goals. Forest landscape restoration is an active, long-term process to regain ecological integrity and enhance human well-being when forest cover, forest qualities and forest-based contributions to people are diminished. Despite the many advances in the development and application of decision support tools in FLR, this review reveals a gap in tools for the implementation of landscape-scale restoration initiatives and for guiding monitoring and adaptive management. The review also reveals that available tools primarily focus on assessing restoration opportunities at a broader scale, rather than within landscapes where implementation occurs. Evidence from research on community-based conservation and forest management suggests that tools for the empowerment, land rights and capacity building of local residents can help nurture strong coalitions of landscape restoration practitioners that apply adaptive management of restoration interventions, and evaluate potential restoration scenarios in their own landscapes.

CIFOR Occasional Papers contain research results that are significant to tropical forest issues. This content has been peer reviewed internally and externally.

Center for International Forestry Research (CIFOR) CIFOR advances human well-being, equity and environmental integrity by conducting innovative research, developing partners’ capacity, and actively engaging in dialogue with all stakeholders to inform policies and practices that affect forests and people. CIFOR is a CGIAR Research Center, and leads the CGIAR Research Program on Forests, Trees and Agroforestry (FTA). Our headquarters are in Bogor, Indonesia, with offices in Nairobi, Kenya, Yaounde, Cameroon, and Lima, Peru.

ISBN 978-602-387-070-7 DOI: 10.17528/cifor/006792

http://cifor.org

decision support tools for forest landscape …decision support tools for forest landscape...

Documents