The purpose of this study was to improve understanding of the key factors that drive the adoption of tree-based systems (TBS) at scale in Rwanda in order to increase the effectiveness of agroforestry interventions designed to help poor rural farmers enhance their food, income, and energy security. The study assessed differences between and within six land use systems across Rwanda, with a mix of survey, mapping, and focus group methodologies. It also proposed an improved pathway for increased adoption at scale of tree-based systems. Some key findings were as follows.

• There is widespread adoption of agroforestry systems in Rwanda, but the intensity of adoption varies greatly.

• There are pronounced differences across the six land use systems in the types of agroforestry systems adopted, and there is considerable variation within each land use system in the degree of adoption.

• Agroforestry adoption was highest where the population is the densest and farm sizes are the smallest. These areas have intensive farm systems that integrate agriculture, livestock, and wood production.

• Essentially all agricultural land in the country is suitable for some kind of agroforestry.

• Eucalyptus is by far the most common tree planted in Rwanda.

• Timber trees, Eucalyptus, and Alnus were the most common species among species integrated within farms (as opposed to woodlots).

• Multipurpose agroforestry species (Grevillea and Calliandra) and fruit trees are found in lower numbers and have quite significant potential for scaling-up.

• Trees are adopted as part of integrated systems along with crop production.

• There is great potential for intercropping fertilizer and fodder shrubs in crop fields and on terraces to enhance soil fertility and crop yields.

• There is a complex array of factors influencing the adoption of agroforestry systems and the differences observed among the various systems. Proximity to extension services, the history of past interventions, and the availability of tree seeds and seedlings were found to be important drivers to accelerate the adoption of agroforestry at a much greater scale.

• The selection of locally appropriate systems and local factors such as market access and education are important aspects.

• In Rwanda, success in accelerating the scaling-up of agroforestry will be most readily achieved through obtaining strong government and nongovernmental organization (NGO) support, developing agroforestry value chains, and inducing more-vibrant partnerships between research and development agencies.

The policy and regulatory framework of the Rwandan government should infuse tree-based ecosystem approaches (TBEAs) into the national extension agenda. External partners and donors could then support the agenda more vigorously, as guided by the Ministry of Natural Resources in collaboration with the Ministry and Agriculture and Animal Resources and the Ministry of Local Government. External expertise could be harnessed more effectively to provide strategic technical backstopping.

The current extension approaches used for food crops—such as mass dissemination, Farmer Field Schools, and innovation platforms—should now be deployed to create a robust TBEA scaling-up pathway. The Rwanda Natural Resources Authority , which includes the directorate of forestry extension, could chair the forum of partners implementing TBEAs, in collaboration with the Rwanda Agriculture Board and the directorate of agricultural extension at zone level. The national budget can be channeled to accelerate the scale of the agroforestry roll-out through earmarked transfers and the Rwanda Agriculture Board extension services. This would help coordinate the current scattered efforts in TBEA extension undertaken by individual NGOs and projects.