Co-producing biodiversity data infrastructures: Mexico’s Agrobiodiversity Information System as a case study

Agrobiodiversity data, rooted in place and in community, expose tensions around sustaining the involvement of data collectors across the data lifecycle and embedding their context-specific expertise as data are mobilized beyond their sites of origin. These challenges extend to biodiversity data infrastructures that integrate diverse socio-ecological information. We examine how a co-production approach to infrastructure design and development can sustain data collectors’ participation beyond data generation, drawing on our experience developing the Agrobiodiversity Information System (SIAgroBD) in Mexico. We analyze three participatory strategies implemented during SIAgroBD’s development: data literacy workshops, collaborative assessment of data collection tools, and co-design of a data management plan. Our analysis draws on workshop materials, facilitation notes, surveys, focus groups, and project documentation. Data literacy workshops helped collectors situate their work within the wider infrastructure, thus creating conditions for participation beyond the data generation stage. Participatory tool assessment surfaced context-specific adaptations and tacit knowledge, which collectors translated into shared documentation. Through scaffolded co-design, context-specific expertise from data collectors informed access and reuse agreements for sensitive data. Involving data collectors did not, on its own, make the development of SIAgroBD a co-production process. Meaningful participation required spaces for knowledge exchange and mechanisms through which their expertise could inform infrastructure development beyond the data collection stage, with different levels of involvement across the data lifecycle. These activities required substantial time, coordination and resources. Capacity building and feedback processes should therefore be recognized as integral components of infrastructure development.