Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 1 Post-print of: Scheidel, A. and S. Gingrich. 2020. Toward sustainable and just forest recovery: 2 research gaps and potentials for knowledge integration. One Earth, 3(6), 680–90. 3 https://doi.org/10.1016/j.oneear.2020.11.005 4 5 Towards sustainable and just forest recovery: research gaps and potentials for knowledge 6 integration 7 8 Arnim Scheidel a,1, Simone Gingrich b,1* a 9 Universitat Autónoma de Barcelona (ICTA-UAB), Institute of Environmental Science and 10 Technology, Bellaterra, Spain. ORCID: 0000-0001-9764-685X b 11 University of Natural Resources and Life Sciences Vienna (BOKU), Department of 12 Economics and Social Sciences, Institute of Social Ecology, Vienna, Austria. ORCID: 0000- 13 0001-7891-8688 1 14 These authors contributed equally. 15 *Correspondence: simone.gingrich@boku.ac.at 16 17 eTOC 18 High hopes are placed on forest recovery for mitigating climate change and benefitting local 19 communities, but severe ecological and social concerns prevail over its impacts on the ground. 20 We propose that further linking two interdisciplinary research fields, land system science and 21 political ecology, helps addressing these concerns. For five knowledge areas we discuss 22 problems related to lacking knowledge integration, identify specific contributions by the two 23 fields, and outline future research directions to advance ecologically sustainable and socially 24 just forest recovery. 25 1 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 26 Summary 27 Forest recovery is central for addressing major sustainability challenges such as climate change 28 and biodiversity loss. While positive assessments prevail over the global ecological forest 29 restoration potential, critical research highlights limited potentials and even detrimental local 30 impacts, particularly in the Global South. Here we argue that knowledge integration across land 31 system science (LSS) and political ecology (PE) can contribute to addressing this contradiction 32 and advance knowledge about ecologically sustainable and socially just forest recovery. We 33 identify five key areas where knowledge integration is promising: (1) developing multi- 34 dimensional forest definitions, (2) linking forest land to users and interests, (3) identifying 35 reforestation failures and successes, (4) associating drivers and impacts across places and 36 scales, and (5) including justice dimensions in assessments of socio-ecological forest recovery 37 potentials. For each knowledge area, we review key contributions by LSS and PE, and outline 38 future research directions to address ecologically sustainable and socially just forest recovery. 39 40 Keywords 41 Land system science, political ecology, transdisciplinarity, justice, conflict, reforestation, 42 afforestation, forest transitions, conservation, climate change mitigation 43 44 2 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 45 Introduction 46 Forest recovery is a central aim for combatting major sustainability challenges such as climate 47 change1 and biodiversity loss2. While deforestation is an ongoing trend in the global tropics3,4, 48 a growing number of studies has identified large reforestation potentials, in particular with 49 regard to their contribution to climate change mitigation.5,6 Forest recovery has also become a 50 policy pillar in global to national climate change mitigation plans, including the United Nations 51 REDD+ program („Reducing Emissions from Deforestation and Forest Degradation”). Tree 52 planting is part of many corporate climate actions7, as well as a key component of sustainability 53 strategies involving nature-based solutions8. Forest conservation is frequently expected to bring 54 co-benefits for local communities. For instance, the Sustainable Development Goals emphasize 55 how forestry measures are an „investment in people and their livelihoods, especially the rural 56 poor, youth and women“ (see https://sustainabledevelopment.un.org/topics/forests). 57 This positive perception on the potential role of forest recovery to benefit people and the planet 58 has been increasingly countered by critical research concerned with the ecological feasibility 59 and the social impacts of large-scale reforestation and forest conservation. Debates on the 60 ecological feasibility revolve around the realistic potential for reforestation in terms of climatic 61 suitability of land areas9 and the expected carbon sequestration potential10. Among the negative 62 social outcomes, land conflicts over industrial tree plantations have been identified across the 63 Global South11. Reforestation initiatives promoted in the context of climate change mitigation 64 policies have been linked to livelihood loss in many national and local contexts12,13. There is 65 also growing concern over the negative social impacts of expanding conservation areas, which 66 frequently target forests, on the livelihoods of customary land users14. 67 We argue that further knowledge integration across land system science (LSS)15 and political 68 ecology (PE)16 has much to offer to enhance a nuanced understanding of sustainable and just 3 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 69 forest recovery processes. Both fields address, in different ways, the complexities of social and 70 environmental factors shaping forest recovery, as pointed out in an extensive review article 71 detailing the explanatory claims of political ecology and land change science, the major 72 precursor of land system science17. LSS has a tradition in mapping land system change (in 73 particular: land use and land cover change), focusing on socio-ecological phenomena such as 74 ecosystem services, land-atmosphere processes, land governance and urban-rural 75 teleconnections.15 Rooted in land change science, many LSS studies adopt a post-positive 76 approach relying on empirical methods, modelling and testing, which creates interfaces to 77 analyses of the natural world, such as biogeochemical fluxes, or biodiversity trends, and, in 78 principle, allows applicability of methods at multiple scales.17 PE on the other hand, rooted in 79 critical social sciences, frequently adopts constructivist or post-marxist perspectives, and 80 focuses on power relations, conflict and justice concerns associated to specific resource uses. 81 PE thus addresses land characteristics such as tenure, access and diverse forms of material and 82 cultural land uses, and establishes interconnections between political processes and 83 environmental outcomes, including conflicts and social injustices related to land-use change.17 84 While the two fields importantly cross-fertilize each other, the distinct viewpoints and 85 epistemologies hamper a straight-forward integration of approaches.17,18 Zimmerer19, for 86 example, describes how the integration of empirical methods into PE may even pose 87 ‘professional risks’, as it may provoke antagonistic reactions by colleagues who are 88 unsupportive of such integrative efforts. 89 Here we argue that further knowledge integration from PE and LSS into forest recovery 90 research is a risk worth taking, because it has the potential to advance knowledge on sustainable 91 and just forest recovery pathways. In fact, integration is already occurring in important areas 92 and ‘hybrid research’ is emerging that, while not reconciling distinct epistemologies, fosters 4 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 93 “an understanding and appreciation of divergent approaches”17. Rather than seeking full 94 integration of knowledge systems, such research acknowledges productive tensions, or “fruitful 95 frictions”19 that provoke mutual learning and guide careful and selective integration of methods 96 and approaches with the purpose to address specific problems in original and meaningful ways. 97 In this Perspective, we identify where further knowledge integration across PE and LSS is 98 promising for enhancing our knowledge about sustainable and just forest recovery pathways. 99 We draw attention to five knowledge areas in which we consider a selective integration of 100 knowledge and methods to be productive (Figure 1), based on emerging contributions at the 101 interface of the two fields, or on identified research gaps: (1) developing multi-dimensional 102 forest definitions, (2) linking forest land to users and interests, (3) identifying failures and 103 successes of forest recovery, (4) associating drivers and impacts across places and scales, and 104 (5) including justice dimensions in assessments of socio-ecological forest recovery potentials. 105 The identification of key areas for knowledge integration was initially informed by our own 106 research experience on forest recovery processes in different interdisciplinary contexts 107 (including LSS and PE), and then refined in an iterative way through a literature review (see 108 Experimental Procedures section). For the discussion of these knowledge areas, we provide a 109 narrative review that (1) describes the problem arising from a lack of knowledge integration, 110 (2) discusses particular contributions to addressing this problem by LSS and PE, and (3) 111 highlights, based on research at the interface of the two fields wherever available, in which 112 concrete ways further knowledge integration appears promising. Given the diversity of PE and 113 LSS research and their sometimes blurred boundaries, the specific contributions highlighted 114 here are not intended to indicate strict topical divisions between the fields, but tendencies and 115 strengths in approaching forest recovery concerns that bear potential for further cross- 116 fertilization. 5 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 117 With regard to social aspects of forest recovery, we focus on justice, conflict, and the role of 118 specific actors involved, i.e. customary groups or social movements. We take a trivalent 119 perspective to justice that considers distributional aspects, such as who benefits from certain 120 land uses, as well as procedural issues, such as involvement in forest governance, and 121 recognition, concerned with how different cultural identities and related forest uses are valued 122 and respected.20,21 The most important ecological dimensions we address relate to major global 123 sustainability challenges, focusing on the carbon sink function of forest recovery connected to 124 mitigating global climate change1, but including also biodiversity conservation2. 125 Geographically, we mainly address forest recovery processes in the Global South. Analyses of 126 deforestation are included only if they draw conclusions for forest recovery or illustrate 127 methodological or conceptual contributions that could be applied to forest recovery research. 128 129 Figure 1: Knowledge areas for sustainable and just forest recovery research. 6 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 130 Develop multidimensional forest definitions 131 The choice of a particular forest definition sensitively shapes ecological and social outcomes 132 of forest recovery efforts on the ground.22–24 International agreements on climate change 133 mitigation refer to forests as minimum areas of 0.5–1 ha with a tree canopy cover of more than 134 10–30%, comprised of trees higher than 2–5m.25 However, this definition does not capture 135 important ecological and social qualities of forests and may obscure declines in carbon stock, 136 biodiversity loss, and livelihood impacts related to forest change. These impacts can be grasped 137 in interdisciplinary approaches, and we argue that knowledge integration from LSS and PE can 138 inform on suitable dimensions for monitoring forest change (Figure 2). 139 140 Figure 2: Knowledge Area 1 - Develop multidimensional forest definitions. 141 142 Several studies from the field of conservation ecology have highlighted how structural forest 143 definitions may foster ecological degradation.26,27 The central concern of these studies is that 7 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 144 policies targeting forest conservation for climate change mitigation do not distinguish between 145 natural forests and plantations, allowing conversions of old-growth forests into monocrop tree 146 plantations that technically remain forests despite severe ecological degradation, such as 147 biodiversity loss. LSS studies have shown that such political reforestation targets resulted in a 148 shift from primary forest to forest plantations in different geographical contexts.28,29 149 Furthermore, the low threshold for forest cover allows for selective logging of up to 70-90% 150 in closed-canopy forests, leading to carbon losses, without being considered deforestation.23 151 These important ecological changes remain invisible in forest assessments based on such 152 structural definitions.30 More nuanced definitions are required that consider diverse ecological 153 aspects, as well as a more holistic perspective that understands forests embedded in both 154 ecological and social landscapes.22 155 Forest definitions also have major social implications: they reflect particular management 156 objectives, social values and needs attached to forests. The structural definitions underlying 157 those applied by the UNFCCC were initially developed to monitor and manage forests as 158 sources of timber, no matter whether they were planted or natural forests. Thus, they reflected 159 the interests of the timber, plantation and pulp industries but not those of other groups, for 160 instance, Indigenous people relying on forests materially and culturally.22 This has provoked 161 social consequences, particularly for forest-dependent communities whose access to livelihood 162 resources and cultural sites is threatened when forests are converted to tree plantations, or for 163 adjunct land users facing adverse social-environmental impacts of tree plantation expansion, 164 such as pollution, wildlife loss and other issues.31 Environmental movements summarized their 165 profound social concerns over the prevailing use of structural forest definitions with the slogan 166 ‘tree plantations are not forests’, and urged the FAO through mass petitions to revise its 167 definition.32 These concerns are exacerbating in the context of climate change mitigation 8 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 168 policies, where several studies document the profound social problems resulting from the 169 accelerated expansion of tree plantations, endorsed by the UNFCCC as reforestation.12,13,33 170 All forest definitions will require practicable compromise, and no single forest definition will 171 be able to capture all socio-ecological aspects. However, a multidimensional conceptualization 172 of forests is needed for management, monitoring and restoration within current mitigation 173 efforts that must be informed by the multiple social and ecological functions that specific forest 174 landscapes provide.22,34 Definitions used in the governance, restoration and monitoring of 175 forest landscapes must be informed by what is at risk.23 Definitions in the global Forest 176 Resource Assessments, distinguishing ‘natural’ and ‘planted’ forests4 offer only a very rough 177 distinction in this direction. Monitoring the carbon sequestration of forest change should move 178 away from binary definitions of very few different types of forest, ‘other wooded land’, and 179 non-forested areas. Such comprehensive assessments are common in LSS. Building on 180 remotely-sensed data, studies develop and operationalize non-binary land classifications, 181 distinguishing different levels of e.g. carbon stocks3,35 or tree cover36,37. 182 For understanding social aspects of forest change, as well as the implications that specific forest 183 definitions have for different user groups, it is necessary to look at who is at risk, why and how, 184 which are issues at the core of PE.16 Attention must be paid to peoples’ access schemes to 185 forests that in turn shape their vulnerability.38 The only social distinction made in national 186 reports of the Forest Resource Assessment is on public vs. private forests. A consistent 187 international distinction between different forms of customary, small-scale and large-scale 188 corporate forest ownership and access rights, partly existing in national forest inventories, 189 would be a first step towards monitoring of social changes and informing policies to avoid 190 ‘social degradation’. Combining social and ecological attributes into a typology of land-use 191 categories suitable for climate change mitigation and biodiversity conservation that is widely 9 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 192 accepted, useful in monitoring and technically also operational, remains a key challenge and 193 research gap. 194 195 Link forest land to users and interests 196 Adequately linking a physical piece of forest land to people using or wanting to use it is a 197 precondition for avoiding conflicts resulting from competing land use claims.39 Several studies 198 report conflicts in the forestry sector that resulted from inadequate forest classifications and 199 improper identification of the actual land users.33,40 Knowledge about who uses which forest 200 land for what is therefore a prerequisite for preventing conflict and developing socially just 201 forest recovery initiatives. Such identification can build on concepts and methods being 202 debated and developed in both LSS and PE (Figure 3). 203 204 Figure 3: Knowledge Area 2 - Link forest land to users and interests. 10 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 205 LSS has a strong tradition in linking land to people, because one of its core interests is in 206 understanding land system dynamics at the interface of biophysical and human subsystems.41 207 Integrated conceps used in LSS like ‘land functions’, denoting the “wide range of goods and 208 services”42 provided by land, or ‘ecosystem services’, describing the “benefits people obtain 209 from ecosystems”43 bridge ecological and socetial dimensions of land systems and, by enabling 210 identification of functions or services relevant to specific actors and interests, offer entry points 211 for addressing issues of justice or conflict in the context of forest recovery. Relevant conceptual 212 contributions from PE on the other hand focus on the diverging interests of different forest 213 users, and the impact of power relations on the potential benefit of forest change for different 214 actors. ‘Green grabbing’ describes the appropriation of resources for environmental ends, with 215 often detrimental effects for local populations.44 This concept is useful to link land cover, such 216 as forests, to specific actors who expect benefits from its protection, while other actors, such 217 as customary land users, get excluded from this land. In this context, the negative impacts of 218 forest restoration programs such as REDD+ projects have been highlighted, they may 219 exacerbate existing inequalities by benefiting political elites while excluding other social 220 groups from access to local resources.45 221 Also empirically and methodologically, both LSS and PE contribute to better tackling the 222 challenges of linking land to people in cases when standard land classifications fail to 223 acknowledge multifunctional forest use. In particular, LSS has contributed to better mapping 224 shifting cultivation based on advanced analyses of remote sensing data, and thus helped 225 identify the importance of forest resources to local livelihoods otherwise not documented.46,47 226 Recently, a proposal to spatially link illicit activities to land uses using remote sensing data 227 was made by Tellmann et al.48, which might be applied to trace e.g. illicit cultivation or logging 228 activities within forests. Several works in PE on the other hand pinpoint that inadequate land 11 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 229 classifications disadvantage moving populations, such as shifting cultivators49 or pastoralists50, 230 as well as the complex livelihood patterns of peasants. Beyond their permanent agricultural 231 plots, peasants depend on frequently unmapped access to forest livelihood resources, such as 232 non-timber forest products (NTFPs) and wildlife from surrounding forests or savannas51. PE 233 unravels that some forest uses, such as NTFPs, are virtually ‘invisible’, but nevertheless vitally 234 important for local livelihoods.52 235 Further knowledge integration across LSS and PE is promising to better understand who uses 236 which forests and in what ways. At local levels, a number of different contributions have 237 recently pushed this frontier: Hunsberger et al.53 proposed the combination of a landscape 238 perspective, collaborative action research and knowledge co-production with local actors to 239 uncover local histories of land uses as well as local understandings of justice. Participatory 240 mapping has proven to be a valuable tool to integrate information on both social and ecological 241 forest characteristics and diverging interests among different users.54,55 Interview-based 242 research also enables identification who uses which land for what purpose, regardless of 243 whether these uses are considered in standard land classifications or detectable by remote 244 sensing methods.56,57 245 Less work in this direction is available at national and global levels, but important progress has 246 recently been made. For example, Leijten et al.58 map the global forest area that may be covered 247 by corporate zero deforestation commitments based on specific biophysical criteria, and 248 consider potential leakage effects. Fa et al.59 integrate Indigenous Land maps with maps of 249 Intact Forest Lands and show that at least 36% of intact forest lands are within Indigenous 250 peoples’ lands.60 They argue that recognition of Indigenous rights is therefore critical to 251 mitigate deforestation. Similarly, Schleicher et al.14 demonstrate that 1 billion people would be 252 affected if half the terrestrial land surface was put under conservation, as proposed by 12 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 253 international conservationists. Combining approaches from PE that uncover multiple interests 254 in forests, with mapping of land uses associated to specific user groups appears as a promising 255 area for future methodological innovation. 256 257 Identify failures and successes in forest recovery 258 The failure and success of concrete reforestation initiatives depends on both ecosystem and 259 social dynamics. Thompson61 illustrates this for the case of mangrove reforestation in Thailand, 260 where failures have occurred due to a mix of ecological factors, such as tidal inundation, algal 261 accumulation, or invasion by barnacles, as well as political and institutional aspects, including 262 how ecological knowledge was used, or how decisions on siting, planting techniques and 263 monitoring have been made, shaped by the different interests and power relations across 264 involved actors. A thorough understanding of reforestation successes and failures must draw 265 on both ecosystem and political-ecological analyses. While literature is available that addresses 266 institutional and stakeholder dynamics in reforestation projects62, as well as the economic 267 preconditions of reforestation63, bridging LSS and PE can further enhance knowledge about 268 the interplay of ecosystem dynamics, socio-economic and institutional processes, and informal 269 and non-institutionalized processes, shaping reforestation failures and successes (Figure 4). 270 The societal processes enabling reforestation are a major topic in LSS, debated as ‘drivers’ or 271 ‘pathways’ of forest transitions. Work here builds on theoretical approaches that discern ideal- 272 type pathways associated to socio-economic or political factors enabling reforestation.64 273 Originating from a focus on long-term studies of forest change in industrialized countries65,66, 274 major socio-economic drivers of forest recovery, such as agricultural intensification67 and 275 growing income or socio-economic wellbeing68 are increasingly documented in quantitative 13 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 276 assessments of reforestation in the Global South69,70. Complementary, case-specific analyses, 277 including more qualitative methods, highlight the influence of political and institutional 278 dimensions.71 In particular, the historical context of forest governance72 and institutional 279 legacies73 have been demonstrated to impact dynamics of forest change. 280 281 Figure 4: Knowledge Area 3 - Identify failures and successes in forest recovery 282 283 PE contributes to the understanding of processes enabling or disabling various forms of 284 reforestation, not only by addressing the social outcomes, such as justice concerns, but also by 285 shedding light on informal actors, power relations and illicit processes frequently overseen. 286 The relevance of hidden agendas, informal and illicit processes in shaping forests on the ground 287 manifests for example in Cambodia, whose recent political history has been deeply entrenched 14 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 288 with the development of the illicit timber sector.74 While the officially demarcated conservation 289 areas contain some of the most valuable timer resources, they are being logged at a rapid pace 290 by political elites involved in the multi-million dollar shadow economy.75 Structural forest 291 definitions further support this process, because they enable Cambodia’s political elites and 292 timber industries to conduct selective logging, and to replace natural forests containing high 293 value timber stocks by tree plantations that are globally accepted as forests.12 Illicit activities 294 and clientelist relations between the state and the logging industry mark forest governance not 295 only in Cambodia, but occur in many countries targeted for reforestation. They must be taken 296 into account to avoid naïve policy recommendations that do not meet local realities.76 297 PE also sheds light on ‘hidden successes’ of forest recovery, by addressing the contributions 298 of customary resource users and non-institutionalized actors such as social movements to forest 299 conservation, which are not officially labelled as conservation or climate change mitigation 300 actions.77 Coined by Martinez-Alier78 as ‘environmentalism of the poor’, the roles of these 301 actors in environmental protection have been highlighted by PE for decades.79 For example, 302 shifting cultivators, frequently blamed by governments and domestic policies for hindering 303 climate change mitigation, can play an important role in sustainably managing tropical 304 forests.80 Social movements in general can be key in promoting and defending sustainable use 305 of commons against outside threats.81 Specifically, forest movements, such as for example the 306 Cambodian Prey Lang Community Network (PLCN), actively protect the forests upon which 307 their livelihoods depend against illegal loggers and outside encroachers.82 Only recently, the 308 UN Human Rights council formally recognized the role of such ‘environmental defenders’ for 309 sustainability.83 310 Combining further the insights from LSS and PE can contribute to a better understanding of 311 both evident and hidden successes and failures of forest conservation policies. The study of 15 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 312 shifting cultivation is a good example of such knowledge integration. Interdisciplinary 313 approaches have illustrated how too narrow perspectives fail to understand the role of shifting 314 cultivation in climate change mitigation84, and have shed light on the conditions under which 315 shifting cultivators are able to contribute to carbon storage and biodiversity conservation80,85. 316 Similarly promising research links Indigenous lands and land use practices with a quantitative 317 analysis of forest change, highlighting the importance of reckognizing indigenous land uses 318 and involving indigenous groups in forest conservation.59,60 319 Within current debates on the role of environmental defenders for sustainability83, further 320 knowledge is needed to better understand the role of forest defenders for forest recovery 321 globally. While PE has provided important answers to the questions of why and how customary 322 users and social movements protect the environment such as their forests, little has been done 323 to actually track and quantify these contributions. LSS methods, such as the analysis of remote 324 sensing data, can help address these questions by unveiling and monitoring historical and 325 current landscape transformations associated to informal processes and non-instituionalized 326 actors that do not appear in any statistics or official reports (see 48). 327 328 Associate drivers and impacts across places and scales 329 Forest recovery in specific places is usually the result of ecological and societal processes 330 interacting across places and scales. One problem arising from this, known as leakage, is that 331 environmental policies targeting a particular country may result in displacing environmental 332 burdens beyond national boundaries. For example, Ingalls et al.86 trace how REDD+, while 333 supporting forest recovery in Vietnam, indirectly displaces deforestation to Laos and 334 Cambodia, by provoking large-scale land deals for forest-risk commodities in these countries. 16 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 335 A related problem is that case-specific knowledge on the social justice implications of 336 particular reforestation policies or practices has not been generalized to global levels. 337 Currently, we observe a certain scale-related bias, where global studies of forest change tend 338 to overlook social justice dimensions, whereas local case studies that address socio-political 339 aspects of forest change focus less on ecological implications. LSS and PE offer approaches to 340 address both problems: to link and quantify reforestation drivers and impacts across places and 341 scales, and to upscale knowledge from single cases to systematic and generalized knowledge 342 (Figure 5). 343 344 Figure 5: Knowledge Area 4 - Associate drivers and impacts across places and scales 345 346 LSS has its roots in understanding spatial patterns and dynamics of land use change at multiple 347 scales.41 A particular strength of LSS is its ability to consistently depict land use and land cover 17 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 348 change across scales, forest recovery being one of them. National-level analyses for several 349 countries and regions, experiencing net-reforestation after long periods of deforestation, have 350 shown spatial patterns of reforestation based on remote-sensing data products.87,88 Ecological 351 characteristics, like the type of forests recovering28, carbon stock changes in forests89,90, and 352 biodiversity effects91 can be addressed with such approaches, and linked to global contributions 353 of forests e.g. to ecosystem carbon storage35 and biodiversity92. 354 Telecouplings, i.e. the “socioeconomic and environmental interactions over distances”93 have 355 become an important topic of forest recovery analysis in LSS in recent years94. The 356 displacement of deforestation beyond national boundaries via imports of wood or forest-risk 357 commodities has been identified as enabling factor for reforestation at national95 and global 358 scales96. Interventions in international supply chains have been shown to impact the slow-down 359 of deforestation in the Amazon97, and similarly, policy interventions reduced the rate of palm- 360 oil expansion in Indonesia98. 361 Telecouplings research has, despite its focus on power relations in land system change, so far 362 not sufficiently addressed issues of social justice, as recently claimed by Corbera et al.99. 363 Similarly, calls for more contextual analyses of national reforestation processes are being 364 made, arguing that biophysical processes of forest change should be better linked to political, 365 institutional or cultural processes, including conflicts from diverging interests and unequal 366 power relations.100–102 367 Conceptually, the distinction of proximate causes and ultimate driving forces, introduced by 368 land change scientists almost two decades ago103, offers important potential here. This 369 distinction has been successfully applied in many studies on deforestation104,105, enabling the 370 localization of different drivers and identification of actors with diverging interests106. 371 Applying this concept to cases of reforestation appears as a promising next step. 18 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 372 Studies from PE have demonstrated that forest recovery projects are embedded in socio- 373 political processes unfolding at different scales and involve actors with diverging agendas. Key 374 research topics include e.g., political implications of the growing attention to tree 375 plantations107, analyses of actors’ interests and interactions across scales108, conflicts playing 376 out between local populations and national or international investors (e.g., in the context of tree 377 plantations11) or national and international governance administrations (e.g., in the context of 378 conservation efforts109). Baird and Barney110 furthermore showed that project impacts go well 379 beyond specific project boundaries and interact with other land uses, by pointing to the 380 cumulative and cross-sectoral impacts of multiple projects in the same landscape, and their 381 implications for customary users. Extensive empirical work has also been conducted analyzing 382 the drivers and (unintended) social effects of reforestation under the REDD+ scheme, where 383 international policies interact with local land uses. Most studies work at local scales and 384 identify structural problems when international restoration schemes meet local realities.111,112 385 The case study approach adopted in many of these studies has yielded ‘thick descriptions’ with 386 high internal validity of place-specific reforestation dynamics and impacts. However, little is 387 usually said about the external validity of the case study and the generalizability of observed 388 processes beyond the specific case. Questions such as how frequent a particular type of forest 389 conflict occurs, or how many people or how much land are affected by specific forestry 390 activities across geographic regions or globally, remain unanswered. 391 Upscaling case-specific insights and creating generalized knowledge is urgently needed to 392 better address issues of social justice and conflict in the context of global forest recovery 393 endeavors. Such generalizations cannot offer the rich contextual analysis of case studies, but 394 unveal general tendencies and patterns with higher external validity. Important contributions 395 towards upscaling case-specific knowledge are made by extensive literature reviews.45,85,113 19 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 396 Beyond this, we observe an increasing number of empirical data compilations established by 397 researchers and non-governmental institutions that can inform future meta-analyses. Examples 398 here include the global Atlas of Environmental Justice (EJAtlas), providing standardized 399 information on environmental conflicts an justice mobilizations worldwide, including some on 400 reforestation and forest conservation programs.79,114 Similarly, the Land Matrix database 401 provides information on international land deals for agriculture, including forestry related 402 projects115, and ID-RECCO116 compiles information on a large number of REDD+ projects. 403 These databases provide the empirical basis for developing functional typologies (e.g., types 404 of conflicts over land, or success criteria of reforestation projects), and enable situating case- 405 specific processes into larger regional and global contexts. 406 Include justice in socio-ecological forest recovery estimates 407 Studies on global forest carbon sequestration117 and the potentials for forest recovery5,8 fuel 408 hopes that forests can substantially contribute to climate change mitigation118. Important 409 empirical and methodological efforts are being conducted to better quantify ongoing trends in 410 forest extent and carbon storage, especially in the tropics where forests continue to decline.3,36 411 Nevertheless, disagreement among existing datasets on biomass stocks35 and among different 412 methodological approaches119,120 prevail. Given the high uncertainty of global forest, tree cover 413 and carbon stock datasets, research quantifying forest recovery potentials face substantial 414 critiques. So far, criticism has addressed in particular the way in which biophysical constraints 415 are considered, affecting potential forest area expansion and the amount of feasible carbon 416 sequestration.9,10,121 417 However, forest recovery is deeply embedded in social processes122, creating also social 418 constraints. A social justice perspective, which recognizes that socially acceptable forms of 20 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 419 reforestation require attention to diverse values attached to forest recovery, as well as to 420 procedural and distributive concerns of multiple users affected by forest recovery processes, is 421 rarely mentioned in assessments of global reforestation potentials.123 The seedlings of a 422 reforestation project will not grow if they are burnt or uprooted in acts of resistance by locals 423 whose lives are negatively affected by the tree plantations (for a review of cases, see11). 424 Livelihood needs, justice concerns, and conflict must therefore be incorporated in forest 425 recovery potential assessments because these factors directly shape the possible forms of, and 426 limits to, forest recovery. Insights from both LSS and PE can contribute to addressing this 427 challenge (Figure 6). 428 429 430 Figure 6: Knowledge Area 5 - Include justicein assessments of socio-ecological forest recovery 431 potentials 21 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 432 433 Methodologically, ecological potentials have been linked to social issues by overlaying maps 434 with ecological and social information. For example, the number of people potentially affected 435 by forest recovery124, or nature conservation in general14, has been assessed this way. Similarly, 436 Garnett et al.60 use maps of Indigenous territories to identify areas that are under traditional 437 management and subject to Indigenous Peoples’ rights to land that must be respected to meet 438 social and ecological conservation goals. Whether forestry initiatives in disputed areas actually 439 lead to conflicts, or are seen by affected actors as beneficial, depends on multiple factors. 440 Among them are the specific project characteristics, historical land uses, competing land 441 claims, worldviews, local perceptions of justice, and also the resulting distribution of benefits 442 and burdens across different actor groups. A PE lens can unpack the role of these factors, and 443 by scaling up knowledge, enhance a global understanding of conflict triggers associated to 444 different types of forest recovery initiatives. Also, debates from LSS might prove as fruitful 445 entry points. For example, the concept ‘competition over land’125,126 can be useful to identify 446 not only biophysical trade-offs, but also diverging interests by different groups of actors, and 447 the social conflicts associated to them. Methodologically, agent-based modelling is a useful 448 tool to operationalize in a formalized way the interests by different actors in the context of land 449 use,127 but has only rarely been applied to conflicts over forest recovery projects128. 450 Insights on the socio-political dynamics associated to different forms of forest recovery are 451 necessary to revise current estimates of global, regional and national reforestation potentials to 452 more realistic ones, and to guide the forms of forest recovery initiatives towards projects that 453 are socially just. For example, a recent study for Southeast Asia that integrated socio-economic 454 aspects estimated that financial, land use and operational constraints reduce the biophysical 455 reforestation potential by 82% to 99.7%.129 Crucial knowledge gaps prevail regarding the 22 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 456 integration of socio-economic aspects at the global level, and particularly, the inclusion of 457 social justice aspects as constraints to forest recovery. While the latter requires to think about 458 new methodological approaches that draw on LSS and PE insights, it is a promising path to 459 unveil more sustainable and just potentials for forest recovery. 460 461 Concluding remarks 462 This Perspective started from the observation that overly positive estimates of forest recovery 463 potentials to mitigate global climate change or biodiversity loss oppose evidence from more 464 critical studies on the ecological constraints and potentially detrimental social impacts of forest 465 conservation, particularly in the Global South. We propose that knowledge integration between 466 LSS and PE offers productive potentials for addressing this contradiction, by (1) developing 467 multi-dimensional forest definitions, (2) adequately linking forest land to users and interests, 468 (3) identifying failures and successes of forest recovery, (4) associating drivers to impacts 469 across places and scales, and (5) including justice dimensions in assessments of socio- 470 ecological forest recovery potentials. In all five knowledge areas substantial knowledge gaps 471 remain that we have outlined here. Yet, contributions are appearing that draw on insights, 472 approaches and methods from LSS, PE and neighboring fields. These studies show a promising 473 direction for further knowledge integration towards well-informed, ecologically sustainable 474 and socially just forest recovery research. 475 With this contribution we aim at stimulating further inter- and transdisciplinary research to 476 close these gaps, even if this requires going beyond established methodologies, well-known 477 routes of scientific collaboration, or even pose ‘professional risks’. Research along the 478 knowledge frontiers sketched out here is in our view indispensable for informing forest 23 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 479 monitoring and governance at multiple levels. We are well aware that the issues addressed here 480 are also profoundly political, and not only academic problems, which cannot be solved by 481 research alone. Nonetheless, the research agenda proposed here can importantly inform 482 negotiation processes around sustainable forest conservation, by making visible major societal 483 and ecological trade-offs and concerns involved in forest recovery processes. 484 485 Experimental Procedures 486 Resource availability 487 Lead contact 488 Further information and requests should be directed to and will be fulfilled by the Lead Contact 489 Simone Gingrich: simone.gingrich@boku.ac.at 490 Materials availability 491 This study did not generate new unique materials. 492 Data and code availability 493 This study did not analyze datasets. 494 Literature review approach 495 This Perspective is based on a review of about 230 publications that informed our arguments. 496 The compilation of relevant research articles was based on three search strategies: articles we 497 knew from our past work, articles from search results in main research databases (Google 498 Scholar, Web of Science) based on keywords related to the five knowledge areas (e.g., 499 keywords related to topical issues, such as “forest definitions,” “forest classification,” 500 “reforestation failure,” “reforestation success,” “drivers forest recovery,” “telecoupling forest” 24 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 501 and others, combined with keywords related to the discussed disciplines, such as “political 502 ecology,” “land change science,” and “land system science”), and handsearched bibliographies 503 of reviewed articles. We selected contributions from LSS and PE based on two criteria: articles 504 that empirically or conceptually illustrate how knowledge integration can be successfully 505 achieved, or articles that discuss social and ecological concerns resulting from insufficient 506 knowledge integration in forest recovery research. Acknowledging that both LSS and PE are 507 heterogeneous, and their boundaries are difficult to trace, we also included studies from 508 neighboring fields, such as economic, physical, and critical geography, conservation ecology, 509 or critical agrarian studies, if they met the above described selection criteria and scope. Due to 510 limitations of space, we selected publications cited here principally according to their topical 511 fit. In cases where several references were available to back up a statement or illustrate an 512 argument, the final selection was based on the criteria of impact (with preference to more 513 frequently cited papers), recency (with preference to more recent publications), and the gender 514 balance in the authors list. 515 516 Acknowledgements 517 S.G. acknowledges funding from the European Research Council (ERC) under the European 518 Union's Horizon 2020 research and innovation programme (ERC StG 'HEFT', GA 757995). 519 A.S. acknowledges funding from the Beatriu de Pinós postdoctoral programme of the 520 Government of Catalonia's Secretariat for Universities and Research of the Ministry of 521 Economy and Knowledge (2017 BP 00023), and support by the ERC project ‘EnvJustice’ (GA 522 695446). The authors wish to thank two highly constructive reviewers for their thorough 523 reading and detailed comments, as well as the editor for their great assistance indeveloping and 524 finalizing the manuscript. 25 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 525 526 Author contributions 527 A.S. and S.G. contributed equally to the conceptualization and writing of the paper. Literature 528 from land system science was mainly reviewed by S.G., while literature from political ecology 529 was mainly reviewed by A.S. 530 531 Declaration of interests 532 The authors declare no competing interests. 533 534 References 535 1. IPCC (2019). Climate Change and Land: an IPCC special report on climate change, 536 desertification, land degradation, sustainable land management, food security, and 537 greenhouse gas fluxes in terrestrial ecosystems P. R. Shukla, J. Skea, E. Calvo Buendia, 538 V. Mansson-Delmotte, H.-O. Pörtner, D. C. Roberts, P. Zhai, R. Slade, S. Connors, R. 539 van Diemen, et al., eds. (In press). 540 2. IPBES (2019). Global assessment report on biodiversity and ecosystem services of the 541 Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services E. 542 S. Brondizio, J. Settele, S. Díaz, and O. Ngo, eds. (IPBES secretariat). 543 3. Baccini, A., Walker, W., Carvalho, L., Farina, M., Sulla-Menashe, D., and Houghton, 544 R.A. (2017). Tropical forests are a net carbon source based on aboveground 545 measurements of gain and loss. Science 358, 230–234. 546 4. Köhl, M., Lasco, R., Cifuentes, M., Jonsson, Ö., Korhonen, K.T., Mundhenk, P., de Jesus 547 Navar, J., and Stinson, G. (2015). Changes in forest production, biomass and carbon: 548 Results from the 2015 UN FAO Global Forest Resource Assessment. For. Ecol. Manag. 549 352, 21–34. 550 5. Bastin, J.-F., Finegold, Y., Garcia, C., Mollicone, D., Rezende, M., Routh, D., Zohner, 551 C.M., and Crowther, T.W. (2019). The global tree restoration potential. Science 365, 76– 552 79. 26 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 553 6. Busch, J., Engelmann, J., Cook-Patton, S.C., Griscom, B.W., Kroeger, T., Possingham, 554 H., and Shyamsundar, P. (2019). Potential for low-cost carbon dioxide removal through 555 tropical reforestation. Nat. Clim. Change 9, 463–466. 556 7. Seymour, F. (2020). Seeing the Forests as well as the (Trillion) Trees in Corporate 557 Climate Strategies. One Earth 2, 390–393. 558 8. Griscom, B.W., Adams, J., Ellis, P.W., Houghton, R.A., Lomax, G., Miteva, D.A., 559 Schlesinger, W.H., Shoch, D., Siikamäki, J.V., Smith, P., et al. (2017). Natural climate 560 solutions. Proc. Natl. Acad. Sci. 114, 11645–11650. 561 9. Bond, W.J., Stevens, N., Midgley, G.F., and Lehmann, C.E.R. (2019). The Trouble with 562 Trees: Afforestation Plans for Africa. Trends Ecol. Evol. 34, 963–965. 563 10. Lewis, S.L., Mitchard, E.T.A., Prentice, C., Maslin, M., and Poulter, B. (2019). Comment 564 on “The global tree restoration potential.” Science 366, eaaz0388. 565 11. Gerber, J.-F. (2011). Conflicts over industrial tree plantations in the South: Who, how and 566 why? Glob. Environ. Change 21, 165–176. 567 12. Scheidel, A., and Work, C. (2018). Forest plantations and climate change discourses: New 568 powers of ‘green’ grabbing in Cambodia. Land Use Policy 77, 9–18. 569 13. Richards, C., and Lyons, K. (2016). The new corporate enclosures: Plantation forestry, 570 carbon markets and the limits of financialised solutions to the climate crisis. Land Use 571 Policy 56, 209–216. 572 14. Schleicher, J., Zaehringer, J.G., Fastré, C., Vira, B., Visconti, P., and Sandbrook, C. 573 (2019). Protecting half of the planet could directly affect over one billion people. Nat. 574 Sustain. 2, 1094–1096. 575 15. Verburg, P.H., Crossman, N., Ellis, E.C., Heinimann, A., Hostert, P., Mertz, O., 576 Nagendra, H., Sikor, T., Erb, K.-H., Golubiewski, N., et al. (2015). Land system science 577 and sustainable development of the earth system: A global land project perspective. 578 Anthropocene 12, 29–41. 579 16. Robbins, P. (2011). Political ecology: A critical introduction (John Wiley & Sons). 580 17. Turner, B.L., and Robbins, P. (2008). Land-Change Science and Political Ecology: 581 Similarities, Differences, and Implications for Sustainability Science. Annu. Rev. 582 Environ. Resour. 33, 295–316. 583 18. Brannstrom, C., and Vadjunec, J.M. Notes for avoiding a missed opportunity in 584 sustainability science: Integrating land change science and political ecology. In Land 585 Change Science, Political Ecology, and Sustainability. Synergies and divergences, C. 586 Brannstrom and J. M. Vadjunec, eds. (Routledge), pp. 1–23. 587 19. Zimmerer, K.S. (2015). Methods and environmental science in political ecology. In The 588 Routledge handbook of political ecology, T. Perreault, G. Bridge, and J. McCarth, eds. 589 (Routledge), pp. 172–190. 27 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 590 20. Schlosberg, D. (2004). Reconceiving Environmental Justice: Global Movements And 591 Political Theories. Environ. Polit. 13, 517–540. 592 21. Martin, A., Coolsaet, B., Corbera, E., Dawson, N.M., Fraser, J.A., Lehmann, I., and 593 Rodriguez, I. (2016). Justice and conservation: The need to incorporate recognition. Biol. 594 Conserv. 197, 254–261. 595 22. Chazdon, R.L., Brancalion, P.H.S., Laestadius, L., Bennett-Curry, A., Buckingham, K., 596 Kumar, C., Moll-Rocek, J., Vieira, I.C.G., and Wilson, S.J. (2016). When is a forest a 597 forest? Forest concepts and definitions in the era of forest and landscape restoration. 598 Ambio 45, 538–550. 599 23. Putz, F.E., and Redford, K.H. (2010). The Importance of Defining ‘Forest’: Tropical 600 Forest Degradation, Deforestation, Long-term Phase Shifts, and Further Transitions: 601 Importance of Defining ‘Forest.’ Biotropica 42, 10–20. 602 24. Van Holt, T., Binford, M.W., Portier, K.M., and Vergara, R. (2016). A stand of trees does 603 not a forest make: Tree plantations and forest transitions. Land Use Policy 56, 147–157. 604 25. UNFCCC (2013). Afforestation and Reforestation Projects under the Clean Development 605 Mechanism (UNFCCC Secretariat). 606 26. Sasaki, N., and Putz, F.E. (2009). Critical need for new definitions of “forest” and “forest 607 degradation” in global climate change agreements. Conserv. Lett. 2, 226–232. 608 27. Van Noordwijk, M., and Minang, P. (2009). If we cannot define it, we cannot save it: 609 forest definitions and REDD. ASB PolicyBrief 15. 610 28. Hua, F., Wang, L., Fisher, B., Zheng, X., Wang, X., Yu, D.W., Tang, Y., Zhu, J., and 611 Wilcove, D.S. (2018). Tree plantations displacing native forests: The nature and drivers 612 of apparent forest recovery on former croplands in Southwestern China from 2000 to 613 2015. Biol. Conserv. 222, 113–124. 614 29. Sloan, S., Meyfroidt, P., Rudel, T.K., Bongers, F., and Chazdon, R. (2019). The forest 615 transformation: Planted tree cover and regional dynamics of tree gains and losses. Glob. 616 Environ. Change 59, 101988. 617 30. Tropek, R., Sedla ek, O., Beck, J., Keil, P., Musilova, Z., imova, I., and Storch, D. (2014). 618 Comment on “High-resolution global maps of 21st-century forest cover change.” Science 619 344, 981–981. 620 31. Carriere, R., and Lohmann, L. (1996). Pulping the South: Industrial tree plantations and 621 the world paper economy (Zed Books). 622 32. WRM (2016). How does the FAO Forest definition harm people and forests? An open 623 letter to the FAO. 624 33. Olwig, M.F., Noe, C., Kangalawe, R., and Luoga, E. (2016). Inverting the moral 625 economy: the case of land acquisitions for forest plantations in Tanzania. Third World Q. 626 36, 2316–2336. 28 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 627 34. Ellison, D., Morris, C.E., Locatelli, B., Sheil, D., Cohen, J., Murdiyarso, D., Gutierrez, 628 V., Noordwijk, M. van, Creed, I.F., Pokorny, J., et al. (2017). Trees, forests and water: 629 Cool insights for a hot world. Glob. Environ. Change 43, 51–61. 630 35. Erb, K.-H., Kastner, T., Plutzar, C., Bais, A.L.S., Carvalhais, N., Fetzel, T., Gingrich, S., 631 Haberl, H., Lauk, C., Niedertscheider, M., et al. (2018). Unexpectedly large impact of 632 forest management and grazing on global vegetation biomass. Nature 553, 73–76. 633 36. Hansen, M.C., Potapov, P.V., Moore, R., Hancher, M., Turubanova, S.A., Tyukavina, A., 634 Thau, D., Stehman, S.V., Goetz, S.J., Loveland, T.R., et al. (2013). High-Resolution 635 Global Maps of 21st-Century Forest Cover Change. Science 342, 850–853. 636 37. Song, X.-P., Hansen, M.C., Stehman, S.V., Potapov, P.V., Tyukavina, A., Vermote, E.F., 637 and Townshend, J.R. (2018). Global land change from 1982 to 2016. Nature 560, 639– 638 643. 639 38. Ribot, J. (2014). Cause and response: vulnerability and climate in the Anthropocene. J. 640 Peasant Stud. 41, 667–705. 641 39. Andriamihaja, O.R., Metz, F., Zaehringer, J.G., Fischer, M., and Messerli, P. (2019). Land 642 Competition under Telecoupling: Distant Actors’ Environmental versus Economic 643 Claims on Land in North-Eastern Madagascar. Sustainability 11, 851. 644 40. Work, C., and Thuon, R. (2017). Inside and outside the maps: mutual accommodation and 645 forest destruction in Cambodia. Can. J. Dev. Stud. Rev. Can. Détudes Dév. 38, 360–377. 646 41. Turner, B.L., Lambin, E.F., and Reenberg, A. (2007). The emergence of land change 647 science for global environmental change and sustainability. Proc. Natl. Acad. Sci. 104, 648 20666–20671. 649 42. Verburg, P.H., van de Steeg, J., Veldkamp, A., and Willemen, L. (2009). From land cover 650 change to land function dynamics: A major challenge to improve land characterization. J. 651 Environ. Manage. 90, 1327–1335. 652 43. Millennium Ecosystem Assessment (2005). Ecosystems and Human Well-being. 653 Synthesis. 654 44. Fairhead, J., Leach, M., and Scoones, I. (2012). Green Grabbing: a new appropriation of 655 nature? J. Peasant Stud. 39, 237–261. 656 45. Barr, C.M., and Sayer, J.A. (2012). The political economy of reforestation and forest 657 restoration in Asia–Pacific: Critical issues for REDD+. Biol. Conserv. 154, 9–19. 658 46. Li, P., Feng, Z., Jiang, L., Liao, C., and Zhang, J. (2014). A Review of Swidden 659 Agriculture in Southeast Asia. Remote Sens. 6, 1654–1683. 660 47. Heinimann, A., Mertz, O., Frolking, S., Egelund Christensen, A., Hurni, K., Sedano, F., 661 Parsons Chini, L., Sahajpal, R., Hansen, M., and Hurtt, G. (2017). A global view of 662 shifting cultivation: Recent, current, and future extent. PLOS ONE 12, e0184479. 29 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 663 48. Tellman, B., Magliocca, N.R., Turner, B.L., and Verburg, P.H. (2020). Understanding the 664 role of illicit transactions in land-change dynamics. Nat. Sustain. 3, 175–181. 665 49. Fox, J., Fujita, Y., Ngidang, D., Peluso, N., Potter, L., Sakuntaladewi, N., Sturgeon, J., 666 and Thomas, D. (2009). Policies, Political-Economy, and Swidden in Southeast Asia. 667 Hum. Ecol. 37, 305–322. 668 50. Lengoiboni, M., Bregt, A.K., and van der Molen, P. (2010). Pastoralism within land 669 administration in Kenya—The missing link. Land Use Policy 27, 579–588. 670 51. Corbera, E., Hunsberger, C., and Vaddhanaphuti, C. (2017). Climate change policies, land 671 grabbing and conflict: perspectives from Southeast Asia. Can. J. Dev. Stud. Rev. Can. 672 Détudes Dév. 38, 297–304. 673 52. Shackleton, C.M., and Pandey, A.K. (2014). Positioning non-timber forest products on 674 the development agenda. For. Policy Econ. 38, 1–7. 675 53. Hunsberger, C., Corbera, E., Borras, S.M., Franco, J.C., Woods, K., Work, C., de la Rosa, 676 R., Eang, V., Herre, R., Kham, S.S., et al. (2017). Climate change mitigation, land 677 grabbing and conflict: towards a landscape-based and collaborative action research 678 agenda. Can. J. Dev. Stud. 38, 305–324. 679 54. Byg, A., Novo, P., Dinato, M., Moges, A., Tefera, T., Balana, B., Woldeamanuel, T., and 680 Black, H. (2017). Trees, soils, and warthogs – Distribution of services and disservices 681 from reforestation areas in southern Ethiopia. For. Policy Econ. 84, 112–119. 682 55. Zaehringer, J.G., Lundsgaard-Hansen, L., Thein, T.T., Llopis, J.C., Tun, N.N., Myint, W., 683 and Schneider, F. (2020). The cash crop boom in southern Myanmar: tracing land use 684 regime shifts through participatory mapping. Ecosyst. People 16, 36–49. 685 56. Carvalho, R., de Aguiar, A.P.D., and Amaral, S. (2020). Diversity of cattle raising systems 686 and its effects over forest regrowth in a core region of cattle production in the Brazilian 687 Amazon. Reg. Environ. Change 20, 44. 688 57. Lhoest, S., Vermeulen, C., Fayolle, A., Jamar, P., Hette, S., Nkodo, A., Maréchal, K., 689 Dufrêne, M., and Meyfroidt, P. (2020). Quantifying the Use of Forest Ecosystem Services 690 by Local Populations in Southeastern Cameroon. Sustainability 12, 2505. 691 58. Leijten, F.C., Sim, S., King, H., and Verburg, P.H. (2020). Which forests could be 692 protected by corporate zero deforestation commitments? A spatial assessment. Environ. 693 Res. Lett. 694 59. Fa, J.E., Watson, J.E., Leiper, I., Potapov, P., Evans, T.D., Burgess, N.D., Molnár, Z., 695 Fernández‐Llamazares, Á., Duncan, T., Wang, S., et al. (2020). Importance of Indigenous 696 Peoples’ lands for the conservation of Intact Forest Landscapes. Front. Ecol. Environ. 18, 697 135–140. 698 60. Garnett, S.T., Burgess, N.D., Fa, J.E., Fernández-Llamazares, Á., Molnár, Z., Robinson, 699 C.J., Watson, J.E.M., Zander, K.K., Austin, B., Brondizio, E.S., et al. (2018). A spatial 700 overview of the global importance of Indigenous lands for conservation. Nat. Sustain. 1, 701 369–374. 30 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 702 61. Thompson, B.S. (2018). The political ecology of mangrove forest restoration in Thailand: 703 Institutional arrangements and power dynamics. Land Use Policy 78, 503–514. 704 62. Lazos-Chavero, E., Zinda, J., Bennett-Curry, A., Balvanera, P., Bloomfield, G., Lindell, 705 C., and Negra, C. (2016). Stakeholders and tropical reforestation: challenges, trade-offs, 706 and strategies in dynamic environments. Biotropica 48, 900–914. 707 63. Wunder, S., Duchelle, A.E., Sassi, C. de, Sills, E.O., Simonet, G., and Sunderlin, W.D. 708 (2020). REDD+ in Theory and Practice: How Lessons From Local Projects Can Inform 709 Jurisdictional Approaches. Front. For. Glob. Change 3, 11. 710 64. Meyfroidt, P., and Lambin, E.F. (2011). Global Forest Transition: Prospects for an End 711 to Deforestation. Annu. Rev. Environ. Resour. 36, 343–371. 712 65. Mather, A.S., Fairbairn, J., and Needle, C.L. (1999). The course and drivers of the forest 713 transition: The case of France. J. Rural Stud. 15, 65–90. 714 66. Rudel, T.K., Coomes, O.T., Moran, E., Achard, F., Angelsen, A., Xu, J., and Lambin, E. 715 (2005). Forest transitions: towards a global understanding of land use change. Glob. 716 Environ. Change 15, 23–31. 717 67. Rudel, T.K., Schneider, L., Uriarte, M., Turner, B.L., DeFries, R., Lawrence, D., 718 Geoghegan, J., Hecht, S., Ickowitz, A., Lambin, E.F., et al. (2009). Agricultural 719 intensification and changes in cultivated areas, 1970-2005. Proc. Natl. Acad. Sci. 106, 720 20675–20680. 721 68. Kauppi, P.E., Sandström, V., and Lipponen, A. (2018). Forest resources of nations in 722 relation to human well-being. PLOS ONE 13, e0196248. 723 69. Ashraf, J., Pandey, R., and de Jong, W. (2017). Assessment of bio-physical, social and 724 economic drivers for forest transition in Asia-Pacific region. For. Policy Econ. 76, 35–44. 725 70. Youn, Y.-C., Choi, J., de Jong, W., Liu, J., Park, M.S., Camacho, L.D., Tachibana, S., 726 Huudung, N.D., Bhojvaid, P.P., Damayanti, E.K., et al. (2016). Conditions of forest 727 transition in Asian countries. For. Policy Econ. 728 71. Riggs, R.A., Langston, J.D., and Sayer, J. (2018). Incorporating governance into forest 729 transition frameworks to understand and influence Cambodia’s forest landscapes. For. 730 Policy Econ. 96, 19–27. 731 72. Singh, M.P., Bhojvaid, P.P., de Jong, W., Ashraf, J., and Reddy, S.R. (2017). Forest 732 transition and socio-economic development in India and their implications for forest 733 transition theory. For. Policy Econ. 76, 65–71. 734 73. Haider, J.L., Neusel, B., Peterson, G.D., and Schlüter, M. (2019). Past management 735 affects success of current joint forestry management institutions in Tajikistan. Environ. 736 Dev. Sustain. 21, 2183–2224. 737 74. Le Billon, P. (2000). The Political Ecology of Transition in Cambodia 1989-1999: War, 738 Peace and Forest Exploitation. Dev. Change 31, 785–805. 31 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 739 75. Milne, S. (2015). Cambodia’s Unofficial Regime of Extraction: Illicit Logging in the 740 Shadow of Transnational Governance and Investment. Crit. Asian Stud. 47, 200–228. 741 76. Delabre, I., Boyd, E., Brockhaus, M., Carton, W., Krause, T., Newell, P., Wong, G.Y., 742 and Zelli, F. (2020). Unearthing the myths of global sustainable forest governance. Glob. 743 Sustain. 3, e16. 744 77. Franco, J.C., and Borras, S.M. (2019). Grey areas in green grabbing: subtle and indirect 745 interconnections between climate change politics and land grabs and their implications 746 for research. Land Use Policy 84, 192–199. 747 78. Martinez-Alier, J. (2003). The Environmentalism of the poor: a study of ecological 748 conflicts and valuation (Edward Elgar Publishing). 749 79. Scheidel, A., Del Bene, D., Liu, J., Navas, G., Mingorría, S., Demaria, F., Avila, S., Roy, 750 B., Ertör, I., Temper, L., et al. (2020). Environmental conflicts and defenders: A global 751 overview. Glob. Environ. Change 63, 102104. 752 80. Fox, J., Castella, J.-C., and Ziegler, A.D. (2014). Swidden, rubber and carbon: Can 753 REDD+ work for people and the environment in Montane Mainland Southeast Asia? 754 Glob. Environ. Change 29, 318–326. 755 81. Villamayor-Tomas, S., and García-López, G. (2018). Social movements as key actors in 756 governing the commons: Evidence from community-based resource management cases 757 across the world. Glob. Environ. Change 53, 114–126. 758 82. Turreira-García, N., Meilby, H., Brofeldt, S., Argyriou, D., and Theilade, I. (2018). Who 759 Wants to Save the Forest? Characterizing Community-Led Monitoring in Prey Lang, 760 Cambodia. Environ. Manage. 61, 1019–1030. 761 83. UN (2019). Human Rights Council resolution 40/L.22. “Recognizing the contribution of 762 environmental human rights defenders to the enjoyment of human rights, environmental 763 protection and sustainable development.” 764 84. Scheidel, A. (2018). Carbon stock indicators: reductionist assessments and contentious 765 policies on land use. J. Peasant Stud., 1–22. 766 85. Dressler, W.H., Wilson, D., Clendenning, J., Cramb, R., Keenan, R., Mahanty, S., Bruun, 767 T.B., Mertz, O., and Lasco, R.D. (2017). The impact of swidden decline on livelihoods 768 and ecosystem services in Southeast Asia: A review of the evidence from 1990 to 2015. 769 Ambio 46, 291–310. 770 86. Ingalls, M.L., Meyfroidt, P., To, P.X., Kenney-Lazar, M., and Epprecht, M. (2018). The 771 transboundary displacement of deforestation under REDD+: Problematic intersections 772 between the trade of forest-risk commodities and land grabbing in the Mekong region. 773 Glob. Environ. Change 50, 255–267. 774 87. Bruggeman, D., Meyfroidt, P., and Lambin, E.F. (2016). Forest cover changes in Bhutan: 775 Revisiting the forest transition. Appl. Geogr. 67, 49–66. 32 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 776 88. Jadin, I., Meyfroidt, P., and Lambin, E.F. (2016). International trade, and land use 777 intensification and spatial reorganization explain Costa Rica’s forest transition. Environ. 778 Res. Lett. 11, 035005. 779 89. Le Noë, J., Matej, S., Magerl, A., Bhan, M., Erb, K., and Gingrich, S. (2020). Modelling 780 and empirical validation of long‐term carbon sequestration in forests (France, 1850‐ 781 2015). Glob. Change Biol., gcb.15004. 782 90. Magerl, A., Le Noe, J., Erb, K., Bhan, M., and Gingrich, S. (2019). A comprehensive 783 data-based assessment of forest ecosystem carbon stocks in the U.S. 1907-2012. Environ. 784 Res. Lett. 14, 125015. 785 91. Otero, I., Marull, J., Tello, E., Diana, G.L., Pons, M., Coll, F., and Boada, M. (2015). 786 Land abandonment, landscape, and biodiversity: questioning the restorative character of 787 the forest transition in the Mediterranean. Ecol. Soc. 20, art7. 788 92. Betts, M.G., Wolf, C., Ripple, W.J., Phalan, B., Millers, K.A., Duarte, A., Butchart, 789 S.H.M., and Levi, T. (2017). Global forest loss disproportionately erodes biodiversity in 790 intact landscapes. Nature 547, 441–444. 791 93. Liu, J., Hull, V., Batistella, M., DeFries, R., Dietz, T., Fu, F., Hertel, T.W., Izaurralde, 792 R.C., Lambin, E.F., Li, S., et al. (2013). Framing Sustainability in a Telecoupled World. 793 Ecol. Soc. 18, art26. 794 94. Friis, C., Østergaard Nielsen, J., Otero, I., Haberl, H., Niewöhner, J., and Hostert, P. 795 (2015). From teleconnection to telecoupling: taking stock of an emerging framework in 796 land system science. J. Land Use Sci., 23. 797 95. Jadin, I., Meyfroidt, P., and Lambin, E.F. (2016). Forest protection and economic 798 development by offshoring wood extraction: Bhutan’s clean development path. Reg. 799 Environ. Change 16, 401–415. 800 96. Pendrill, F., Persson, U.M., Godar, J., and Kastner, T. (2019). Deforestation displaced: 801 trade in forest-risk commodities and the prospects for a global forest transition. Environ. 802 Res. Lett. 14, 055003. 803 97. Nepstad, D., McGrath, D., Stickler, C., Alencar, A., Azevedo, A., Swette, B., Bezerra, T., 804 DiGiano, M., Shimada, J., Seroa da Motta, R., et al. (2014). Slowing Amazon 805 deforestation through public policy and interventions in beef and soy supply chains. 806 Science 344, 1118–1123. 807 98. Busch, J., Ferretti-Gallon, K., Engelmann, J., Wright, M., Austin, K.G., Stolle, F., 808 Turubanova, S., Potapov, P.V., Margono, B., Hansen, M.C., et al. (2015). Reductions in 809 emissions from deforestation from Indonesia’s moratorium on new oil palm, timber, and 810 logging concessions. Proc. Natl. Acad. Sci. 112, 1328–1333. 811 99. Corbera, E., Busck-Lumholt, L.M., Mempel, F., and Rodríguez-Labajos, B. (2019). 812 Environmental Justice in Telecoupling Research. In Telecoupling, C. Friis and J. Ø. 813 Nielsen, eds. (Springer International Publishing), pp. 213–232. 33 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 814 100. Gingrich, S., Lauk, C., Niedertscheider, M., Pichler, M., Schaffartzik, A., Schmid, M., 815 Magerl, A., Le Noë, J., Bhan, M., and Erb, K. (2019). Hidden emissions of forest 816 transitions: a socio-ecological reading of forest change. Curr. Opin. Environ. Sustain. 38, 817 14–21. 818 101. Kull, C.A. (2017). Forest transitions: a new conceptual scheme. Geogr. Helvetica 72, 819 465–474. 820 102. Roy Chowdhury, R., and Moran, E.F. (2012). Turning the curve: A critical review of 821 Kuznets approaches. Appl. Geogr. 32, 3–11. 822 103. Geist, H.J., and Lambin, E.F. (2002). Proximate Causes and Underlying Driving Forces 823 of Tropical Deforestation. BioScience 52, 143. 824 104. Bebbington, A.J., Humphreys Bebbington, D., Sauls, L.A., Rogan, J., Agrawal, S., 825 Gamboa, C., Imhof, A., Johnson, K., Rosa, H., Royo, A., et al. (2018). Resource 826 extraction and infrastructure threaten forest cover and community rights. Proc. Natl. 827 Acad. Sci. 115, 13164–13173. 828 105. Junquera, V., Meyfroidt, P., Sun, Z., Latthachack, P., and Grêt-Regamey, A. (2020). From 829 global drivers to local land-use change: understanding the northern Laos rubber boom. 830 Environ. Sci. Policy 109, 103–115. 831 106. Curtis, P.G., Slay, C.M., Harris, N.L., Tyukavina, A., and Hansen, M.C. (2018). 832 Classifying drivers of global forest loss. Science 361, 1108–1111. 833 107. Kröger, M. (2016). The political economy of ‘flex trees’: a preliminary analysis. J. 834 Peasant Stud. 43, 886–909. 835 108. Seki, Y. (2001). The Political Ecology of the Philippine Reforestation Program: ODA, 836 Government, and Local People. Philipp. Polit. Sci. J. 22, 79–96. 837 109. Redpath, S.M., Young, J., Evely, A., Adams, W.M., Sutherland, W.J., Whitehouse, A., 838 Amar, A., Lambert, R.A., Linnell, J.D.C., Watt, A., et al. (2013). Understanding and 839 managing conservation conflicts. Trends Ecol. Evol. 28, 100–109. 840 110. Baird, I.G., and Barney, K. (2017). The political ecology of cross-sectoral cumulative 841 impacts: modern landscapes, large hydropower dams and industrial tree plantations in 842 Laos and Cambodia. J. Peasant Stud. 44, 769–795. 843 111. Sikor, T., and Cầm, H. (2016). REDD+ on the rocks? Conflict Over Forest and Politics 844 of Justice in Vietnam. Hum. Ecol. 44, 217–227. 845 112. Massarella, K., Sallu, S.M., Ensor, J.E., and Marchant, R. (2018). REDD+, hype, hope 846 and disappointment: The dynamics of expectations in conservation and development pilot 847 projects. World Dev. 109, 375–385. 848 113. Ziegler, A.D., Phelps, J., Yuen, J.Q., Webb, E.L., Lawrence, D., Fox, J.M., Bruun, T.B., 849 Leisz, S.J., Ryan, C.M., Dressler, W., et al. (2012). Carbon outcomes of major land-cover 850 transitions in SE Asia: great uncertainties and REDD+ policy implications. Glob. Change 851 Biol. 18, 3087–3099. 34 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 852 114. Temper, L., Demaria, F., Scheidel, A., Del Bene, D., and Martinez-Alier, J. (2018). The 853 Global Environmental Justice Atlas (EJAtlas): ecological distribution conflicts as forces 854 for sustainability. Sustain. Sci. 13, 573–584. 855 115. Nolte, K., Giger, M., and Chamberlain, W. (2016). International Land Deals for 856 Agriculture. Fresh insights from the Land Matrix: Analytical Report II. 857 116. Simonet, G., Agrawal, A., Bénédet, F., Cromberg, M., de Perthuis, C., Haggard, D., 858 Jansen, N., Karsenty, A., Liang, W., Newton, P., et al. (2018). ID-RECCO, international 859 database on REDD+ projects and programs, linking economic, carbon and communities 860 data (Version 3.0). http://www.reddprojectsdatabase.org. 861 117. Pan, Y., Birdsey, R.A., Fang, J., Houghton, R., Kauppi, P.E., Kurz, W.A., Phillips, O.L., 862 Shvidenko, A., Lewis, S.L., Canadell, J.G., et al. (2011). A Large and Persistent Carbon 863 Sink in the World’s Forests. Science 333, 988–993. 864 118. Lewis, S.L., Wheeler, C.E., Mitchard, E.T.A., and Koch, A. (2019). Restoring natural 865 forests is the best way to remove atmospheric carbon. Nature 568, 25–28. 866 119. Hansis, E., Davis, S.J., and Pongratz, J. (2015). Relevance of methodological choices for 867 accounting of land use change carbon fluxes. Glob. Biogeochem. Cycles 29, 1230–1246. 868 120. Mitchard, E.T., Saatchi, S.S., Baccini, A., Asner, G.P., Goetz, S.J., Harris, N.L., and 869 Brown, S. (2013). Uncertainty in the spatial distribution of tropical forest biomass: a 870 comparison of pan-tropical maps. Carbon Balance Manag. 8, 10. 871 121. Veldman, J.W., Aleman, J.C., Alvarado, S.T., Anderson, T.M., Archibald, S., Bond, W.J., 872 Boutton, T.W., Buchmann, N., Buisson, E., Canadell, J.G., et al. (2019). Comment on 873 “The global tree restoration potential.” Science 366, eaay7976. 874 122. Garcia, C.A., Savilaakso, S., Verburg, R.W., Gutierrez, V., Wilson, S.J., Krug, C.B., 875 Sassen, M., Robinson, B.E., Moersberger, H., Naimi, B., et al. (2020). The Global Forest 876 Transition as a Human Affair. One Earth 2, 417–428. 877 123. Holl, K.D., and Brancalion, P.H.S. (2020). Tree planting is not a simple solution. Science 878 368, 580–581. 879 124. Luedeling, E., Börner, J., Amelung, W., Schiffers, K., Shepherd, K., and Rosenstock, T. 880 (2019). Forest restoration: Overlooked constraints. Science 366, 315–315. 881 125. Niewöhner, J., Bruns, A., Haberl, H., Hostert, P., Krueger, T., Lauk, C., Lutz, J., Müller, 882 D., and Nielsen, J.Ø. (2016). Land Use Competition: Ecological, Economic and Social 883 Perspectives. In Land Use Competition, J. Niewöhner, A. Bruns, P. Hostert, T. Krueger, 884 J. Ø. Nielsen, H. Haberl, C. Lauk, J. Lutz, and D. Müller, eds. (Springer International 885 Publishing), pp. 1–17. 886 126. Smith, P., Gregory, P.J., van Vuuren, D., Obersteiner, M., Havlík, P., Rounsevell, M., 887 Woods, J., Stehfest, E., and Bellarby, J. (2010). Competition for land. Philos. Trans. R. 888 Soc. B Biol. Sci. 365, 2941–2957. 35 Post-print of: https://doi.org/10.1016/j.oneear.2020.11.005 889 127. Matthews, R.B., Gilbert, N.G., Roach, A., Polhill, J.G., and Gotts, N.M. (2007). Agent- 890 based land-use models: a review of applications. Landsc. Ecol. 22, 1447–1459. 891 128. Dumrongrojwatthana, P., Le Page, C., Gajaseni, N., and Trébuil, G. (2011). Co- 892 constructing an agent-based model to mediate land use conflict between herders and 893 foresters in northern Thailand. J. Land Use Sci. 6, 101–120. 894 129. Zeng, Y., Sarira, T.V., Carrasco, L.R., Chong, K.Y., Friess, D.A., Lee, J.S.H., Taillardat, 895 P., Worthington, T.A., Zhang, Y., and Koh, L.P. (2020). Economic and social constraints 896 on reforestation for climate mitigation in Southeast Asia. Nat. Clim. Change 10, 842–844. 36