Introduction

Land management strategies such as agroforestry systems (AFS) have been recognized as vehicles to support the socioecological transitions towards sustainability. Different approaches to upscale their benefits have mainly promoted a restricted set of technological packages, which have often not spread widely beyond some farms or plots because they fail to consider fine-scale variation in the local circumstances within which AFS need to be adapted. Under this scenario, it is crucial to consider that AFS are knowledge-intensive innovations. Therefore, their upscaling requires crafting usable knowledge and developing innovative strategies to produce, reproduce and adapt AFS knowledge that considers the varying biophysical conditions of sites, the specific socio-economic needs, and the local knowledge and perceptions.

Objective

The objective of this paper is twofold. On the one hand, we examine how actors from different levels and sectors are interacting to produce and reproduce AFS knowledge across two contrasting regions. On the other hand, we explore an innovative way to assess local knowledge production and reproduction in the same regions.

Methods

This study analyzed cacao agroforestry systems (CAFS) since they are multi-species systems that could be adapted to different biophysical conditions and socio-economic needs by developing different management options with varying permanent or temporary crops and shade trees. Data collection was based on semi-structured interviews with cacao value chain actors and in-depth interviews with cacao farmers in Cesar and Caquetá, Colombia. Data analysis included a social network analysis to disentangle the relationships between diverse cacao actors and cognitive mapping to assess local knowledge of cacao farmers.

Results

Although the two regional CAFS knowledge networks show a high density, most of the linkages are created because of time-limited projects, which affect the process of knowledge production and reproduction. One of the most important actors of these networks corresponds to farmer associations, which could take the role of facilitator of knowledge and information reproduction. Cognitive mapping reveals the lack of technical assistance in the two regions since there is not a homogenization of the most common management practices of cacao crops. But it also shows some local knowledge related to pest and soil management as well as plant material selection that could be potentially used when scaling CAFS.

Implications

This comparative approach helps to generalize insights on the role of knowledge networks and local knowledge on upscaling knowledge-intensive innovations like CAFS in regions with highly diverse landscapes.