Background: Forests provide an essential resource to the livelihoods of an estimated 20% of the global population. The contribution of forest ecosystems and forest-based resources to poverty reduction is increasingly emphasized in international policy discourse and conservation and development investments. However, evidence measuring the effect of forest-based activities on poverty outcomes remains scattered and unclear. Lack of systematic understanding of forest-poverty relationships, in turn, inhibits research, policymaking, and efficient financial resource allocation. Methods: To identify relevant studies for inclusion in this systematic map we searched six bibliographic databases, 15 organizational websites, eight systematic evidence syntheses (reviews and maps), and solicited information from key informants. Search results were screened for relevance against predefined inclusion criteria at title, abstract, and full text levels, according to a published protocol. Included articles were coded using a predefined framework. Trends in the evidence, knowledge gaps and relatively well-researched sub-topics are reported in a narrative synthesis. Occurrence and extent of existing evidence about links between interventions and outcomes are presented in a visual heatmap. Data are available through the open access Evidence for Nature and People Data Portal (http://www.natureandpeopleevidence.org). Results: A total of 242 articles were included in the systematic map database. Included articles measured effects of 14 forest-based intervention types on 11 poverty dimensions. The majority of the evidence base (72%) examined links between productivity-enhancement strategies (e.g. forest management, agroforestry, and habitat management) and monetary income and/or social capital outcomes. Other areas with high occurrence of articles include linkages between interventions involving governance, individual rights/empowerment or linked enterprises/livelihood alternatives with impacts on monetary income from direct sale of goods. A key knowledge gap was on the impacts of investment-based interventions (i.e. enhancing produced, human, and social capitals). Another was the impacts of forest-based interventions on financial capital (savings, debt), non-monetary benefits, and health. Conclusions: The evidence base on forest-based productive activities and poverty alleviation is growing but displays a number of biases in the distribution of articles on key linkages. Priorities for future systematic reviews and evaluations include in-depth examinations into the impacts of rights-based activities (e.g. governance, empowerment) on poverty dimensions; and productivity-enhancing activities on social capital. More comprehensive and robust evidence is needed to better understand the synergies and trade-offs among the different objectives of forest conservation and management and variation in outcomes for different social groups in different social-ecological contexts.

Certification by the Forest Stewardship Council (FSC) remains rare among forest management units (FMUs) in natural tropical forests, presenting a challenge for impact evaluation. We demonstrate application of the synthetic control method (SCM) to evaluate the impact of FSC certification on a single FMU in each of three tropical forest landscapes. Specifically, we estimate causal effects on tree cover change from the year of certification to 2012 using SCM and open-access, pan-tropical datasets. We demonstrate that it is possible to construct synthetic controls, or weighted combinations of non-certified FMUs, that followed the same path of tree cover change as the certified FMUs before certification. By using these synthetic controls to measure counterfactual tree cover change after certification, we find that certification reduced tree cover loss in the most recent year (2012) in all three landscapes. However, placebo tests show that in one case, this effect was not significant, and in another case, it followed several years in which certification had the opposite effect (increasing tree cover loss). We conclude that SCM has promise for identifying temporally varying impacts of small-N interventions on land use and land cover change.