**1. Introduction**

Land degradation is often exacerbated by climate change, threatening multiple socio-ecological systems and the livelihoods of people whose lives depend on them. Ecological restoration and rehabilitation projects are pursued with a view to mitigating the impacts of climate change by storing carbon in plants and soil, while often reducing people's vulnerability and increasing their capacity to adapt [1]. The central role of trees in the carbon cycle has led to innovative international forest governance mechanisms that seek to recognize the value of forests, in addition to their supply of timber, labor, biodiversity and water as ecosystem services [2]. In line with Article 5 of the 2015 Paris Agreement, the pursuit of results-based payments—e.g., through carbon markets and the Reducing Emissions from Deforestation and Degradation (REDD+) framework—provide financial incentives to governments and communities in the developing world to maintain and increase forest cover [3,4]. It is

acknowledged that these mechanisms o ffer potential to deliver "triple-wins" or climate-compatible development, assisting climate change mitigation, but also reducing vulnerability, alleviating poverty and enabling people to adapt [5].

The capacity to exploit synergies between the multiple benefits delivered by trees and multifunctional landscapes through ecological restoration and rehabilitation has been limited to date by spatial and temporal constraints. Spatially, mitigation benefits are typically global, while adaptation benefits are localized [6]. Temporally, mitigation requires long-term commitment, while adaptation can happen more quickly [7], through accumulated coping strategies and in the longer term, through transformation. While multiple forest and landscape restoration benefits such as carbon sequestration, rehabilitation of degraded lands, flood risk reduction and enhanced biodiversity, can be generated in synergy, trade-o ffs occur when long-term impacts on vulnerability and adaptation capacity are taken into account [8]. For example, a short-term adaptation initiative aimed at building a seawall to protect the community from cyclones and flooding, might attract people to an area perceived as being safe, while it further exposes them to extreme climatic events [9]. Long-term increased vulnerability, or decreased adaptive capacity, deriving from short-term adaptation actions that produce unintended impacts, are defined as "maladaptation" [9].

Adaptation to climate change requires a long-term vision to shift socio-ecological systems towards sustainable trajectories. As identified by Cinner et al. (2018) [10], there are five key domains involved in building long-term adaptive capacity. These include the financial and technological assets accessed by people in time of need, the flexibility to switch across adaptation strategies, the ability of society to organize and share knowledge, the capacity to learn and the agency to determine whether a change is needed. In light of spatio-temporal considerations, it is imperative to plan and evaluate mitigation and adaptation actions simultaneously, with a view to targeting and addressing multiple scales and timeframes to e ffectively manage climate change impacts [11].

Pursuing long-term mitigation and adaptation is imperative to sustain the livelihoods of natural-resource dependent communities that are a ffected by climate change [12]. Since the mid-2000s increasing focus has been placed on the need to mainstream restoration actions into the development trajectories of developing countries [13]. By reconciling adaptation, mitigation and development, climate change benefits can be combined with poverty reduction, health improvements and increased food security [14]. Using multiple-benefit approaches is of particular relevance in Southern African Development Community nations, where socio-economic development opportunities are jeopardized by high climatic uncertainty [15]. Despite growing literature on the potential of ecological restoration approaches to generate an array of benefits, there are still several gaps that require further empirical case study investigations [11].

Suckall et al. (2015) [16] report that where joint ecological restoration or rehabilitation and climate appraisals have been attempted, di fficulties remain in understanding how well co-benefits map onto the development and climate adaptation challenges perceived by local communities. Analyzing two community-level projects that pursue ecosystem rehabilitation in South Africa, Favretto et al. (2018) [17] observed mismatches in reported and perceived benefits at project and community levels. As noted by Cinner et al. (2018) [10], limited capacity to integrate local knowledge and perspectives into science, policy and practice, coupled with limited participation of local communities in adaptive co-management, constrain the delivery of long-term benefits. An improved understanding of community perceptions is key to allow meaningful community engagement, which is a pre-requisite for projects to deliver on expected outcomes [18]. As di fferent stakeholder groups have di fferent views with respect to mitigation, adaptation and development priorities, understanding how best to assess and integrate local-level perspectives into climate and development policy and practice is key if actions are to secure local buy-in and meet multiple goals.

Madagascar is one of the world's major biodiversity hotspots, presenting an important study context where the gaps in assessing multiple benefits and mainstreaming community priorities into restoration require urgen<sup>t</sup> attention [19]. Forest policy, projects and research in Madagascar

have all historically focused on biodiversity conservation [20], with attention to climate change and development impacts gaining momentum only after REDD+ mechanisms and climate action emerged onto international and national agendas [5]. Analyzing thirty years of conservation e fforts, Waeber et al. (2016) [21] showed that this limited scope of analysis hampers the assessment of progress towards international development goals.

This paper targets the gap noted above, i.e., the need to improve understanding of perceived local co-benefits of ecological restoration across climate and development dimensions with a view to enhance capacity to deliver long-term benefits through integration of local knowledge into practice. It aims to examine perceptions of the extent to which forest projects that pursue ecological restoration and rehabilitation in Madagascar can deliver adaptation, mitigation and development co-benefits to local communities. We focus on two case study projects, based on mangrove reforestation and REDD+ conservation. Such an integrated analysis that links local- and project-level perspectives of climate-development impacts of conservation projects in Madagascar, combining multiple indicators of perceived co-benefits, has not been carried out before. We ask the following research questions:


If land degradation and deforestation continue at current rates, this is likely to result in a range of uncontrolled societal, economic and environmental ramifications which hamper the achievement of international agreements and the Sustainable Development Goals (SDGs) [1]. Lessons from this research are relevant across sub-Saharan Africa, as countries seek to progress towards the SDGs and their Paris Climate Agreement Nationally Determined Contributions (NDCs) [22]. Moving towards the United Nations Decade on Ecosystem Restoration 2021–2030, understanding the prerequisites and challenges for multiple benefits to be achieved in climate and development policy and practice, especially those pursuing job creation and ecological restoration and rehabilitation, is more pertinent than ever [23].

#### **2. Materials and Methods**

#### *2.1. Study Country Context*

Madagascar is a biodiversity hotspot and conservation priority area, as well as one of the world's poorest countries [19]. Between 1953 and 2014, 44% of forest was lost nationally [5]. In 2014, forest cover accounted for 8.9 million ha, with an annual deforestation rate of 1.1% reported in the period 2010–2014 (i.e., 99,000 ha/year) [24]. Increasing temperatures impact on temporal and spatial rainfall distribution and exacerbate drought and flood intensity across the southern Africa region [25]. To adapt to climatic changes and cope with extended droughts, subsistence farmers in Madagascar increased forest clearing by up to 17% between 2000 and 2013, compared to pre-2000 levels [26].

Climate change impacts are exacerbated by political instability, which, combined with geopolitical approaches that internationalize conservation, result in undervaluing local knowledge and perspectives [27]. Since the 1960s, policy and funding e fforts to address the most pressing environmental challenges in Madagascar have focused on biodiversity conservation, with US\$ 700 million of support between 1990 and 2012 [21]. Since 1996, through institution of a system that transfers forest managemen<sup>t</sup> to communities ("Gelose" law), forest managemen<sup>t</sup> decisions have been increasingly transferred from governmen<sup>t</sup> to grassroots community-organizations, and community-based conservation has been put at the center of environmental policy and forest managemen<sup>t</sup> [27]. The 2003 Durban Accord boosted a decentralized approach to protected areas by emphasizing co-managemen<sup>t</sup> in biodiversity conservation to achieve poverty reduction [28,29]. Consequently, between 2003 and 2016, protected areas quadrupled and five carbon-rich biomes have been set up through REDD+ projects [5,19]. Madagascar's climate change policy, in particular its National Adaptation Program of Action (NAPA) and its Nationally Determined Contribution (NDC), highlight large-scale reforestation and local community engagemen<sup>t</sup> to protect the remaining forests. Through such engagement, these policies seek to both promote socio-economic development and enhance climate change mitigation and adaptation [30,31].

#### *2.2. Case Study Project Selection*

Two case study projects that pursue ecological restoration and rehabilitation were selected to investigate their climate-development co-benefits, challenges and trade-offs at the community-level (Figure 1). These two projects were selected because they: (i) pursue ecological restoration and rehabilitation with implications for climate adaptation and mitigation, (ii) cover multiple types of forest governance (protected areas and community forest management) in both coastal and inland settings, (iii) involve project target beneficiaries who are highly vulnerable to climate change and have high dependency on the forest for their livelihoods, (iv) have potential to access reduced emissions forestry schemes such as REDD+ and carbon markets, and (v) have a comparable project implementation history of approximately ten years. A total of six projects relevant to the study theme were initially identified. Only two of them met all of the five selection criteria and were therefore chosen as case studies.

**Figure 1.** Forest restoration and rehabilitation case study projects, Madagascar. Source: Adapted from D-Maps, 2018 [32].

#### 2.2.1. Case Study 1: Eden Projects

Eden is a non-profit organization whose mission is to improve the livelihoods of impoverished Malagasy communities by implementing reforestation and environmental stewardship projects. It operates in the Boeny region, which has an average annual temperature of 27 ◦C and rainfall ranging between 1000 and 1200 mm. Between 1990 and 2013 roughly 7500 ha of forest were lost in Boeny, with average carbon storage reductions of 2.7 million tons/year, mostly due to illegal logging for construction material and charcoal production [33]. Since 2008, Eden has sustained mangrove reforestation with poor communities heavily dependent on coastal mangroves for fishing. Funded mostly by businesses, Eden's development model employs locals to plant trees, with the aim to enhance income, increase adaptive capacity and stop deforestation, moving towards long-term improved mitigation [34].

#### 2.2.2. Case Study 2: Makira Forest REDD+ Project

Since 2012, the Wildlife Conservation Society (WCS) has been managing 372,000 ha of the Makira Forest protected area on behalf of the Madagascar Ministry of Environment and Forests. Makira is one of the most intact and highly biodiverse remaining rainforest systems in Madagascar, where increasing deforestation has been driven by slash and burn rice agriculture and illegal logging, exacerbated by rapid population growth and poverty [35]. The Makira forest REDD+ project, implemented by WCS since 2005, aims to safeguard the Makira Natural Park (in Analanjirofo, Sava and Sofia regions) and contribute towards national emissions reduction targets. It seeks to align avoided deforestation with ecosystem conservation, while supporting the socio-economic development of communities living around the forest. Makira's implementation model involves: (i) developing contracts to transfer forest resources managemen<sup>t</sup> to local communities adjacent to the protected area, (ii) strengthening local institutions and land use planning, and (iii) diversifying income through improved agricultural techniques, infrastructure development, conservation enterprise activities and ecotourism. By selling 'Verified Carbon Standard' (VCS) and 'Climate, Community and Biodiversity Alliance' (CCBA) verified carbon credits generated through avoided deforestation, the project aims to market over 31 million tons of carbon o ffsets to sustain long-term conservation over the period 2005–2034 [35].
