Scaling on-Farm Experiments to Achieve Rapid Localized Climate Adaptation.

Regional farming systems are characterized by high levels of spatial and temporal variation which may be intensified by anomalous events associated with climate change.  On-farm experimentation provides a means to adapt to this uncertainty by identifying localized bioclimatic crop responses and optimal agronomic management.  Unfortunately, the magnitude of interacting factors and required level of experimentation to achieve site-specific management is prohibitive at the single-farm scale.   Instead, coordinating experimentation across multiple similar farms provides a way to transcend these difficulties and simultaneously cultivate a culture of inquiry among groups of farmers.  Presented here are results from several years of on-farm experiments that highlight the statistical and logistical trade-offs between single and multiple-farms.   The data highlight a ‘worst-case scenario’ of heterogeneity in the highly variable environment of Montana, which suggests the optimal level of experimental aggregation as a function of spatial variability and the ideal complexity of the system.  To mitigate the future impacts of climate change over large scales of heterogeneous landscapes, an automated procedure for performing experimentation is proposed.