Prototyping and farm system modelling--Partners on the road towards more sustainable farm systems

Farm system modelling and prototyping are two research methods proposed to enhance the process of developing sustainable farm systems. Farm system models provide means to formalize, expand and refine expert knowledge and to integrate this with scientific agro-ecological knowledge at the farm level. The prototyping methodology was developed for the design of more sustainable farm systems, either on experimental or commercial farms. The main features of prototyping are: (1) quantification of goals; (2) emphasis on multiple societal goals; (3) designing as an organizing principle; (4) iteration of system analysis, design and on-farm testing. Hypothetically, farm system modelling could enrich the prototyping methodology and vice versa. Taking a goal-oriented stance, a modelling exercise could reveal design options otherwise overlooked and extrapolation of prototyping results to other conditions and scenarios. The on-farm prototyping work could serve as a source of inspiration and information for farm system modelers. However, little cross-pollination between the modelling and prototyping efforts has occurred, even though the methodologies have been applied in parallel and in one country. Existing reports on prototyping projects merely present their methodological set-up and results, but lack description of the implementation of the methodology. We deemed insight into the implementation of prototyping essential to understand the discrepancy between theory and practice and to investigate the potential for cross-pollination between modelling and prototyping in the future. Three promising leads were identified to assess this potential, i.e. (1) exploring goals of farm systems; (2) exploring options for a change and improvement of farm systems; (3) communication and extrapolation of project output. Analysis of more than two decades of Dutch prototyping research both on experimental and commercial farms indicated that prototyping on commercial farms is a highly localized process. Moreover, although the methodology manual suggests differently, goal formulation was not a distinctive phase of prototyping on commercial farms, so cross-pollination with farm system modelling could not occur (lead 1). As the timely operationalization and the localization of a farm system model demand considerable effort, contributions of farm model explorations to the localized change process on commercial farms (lead 2) seem impractical and unlikely. For communication and extrapolation of prototyping output (lead 3), issue-specific (i.e. focus on a component of the system) models are increasingly used. For this purpose, we hypothesize that there may also be a role for farm system models.