Investigating transitions to regenerative agriculture using the InSEEDS World-Earth model – foundations, first results, and research directions

Worldwide, developments towards highly industrialized, conventional agriculture systems have led to ecological deterioration, as well as societal problems. On the ecological side, many Planetary Boundary transgressions can - in substantial parts - be attributed to agricultural developments: Land System Change, Biosphere Integrity, Climate Change, Freshwater Change, and Biogeochemical Flows were all found to be majorly impacted by agriculture. Social issues like worker health and unstable livelihoods add to the dire picture.

Many approaches towards sustainable agriculture exist, among them Regenerative Agriculture (RA). While RA practices, according to most definitions, center on improving soil health, broader interpretations of the term exist. Some include factors like agrobiodiversity and water cycles, others additionally include social aspects like farmer well-being. While studies assessing the biophysical potential of different RA practices exist, the question how a wide-spread, up-to-global adoption of RA practices can unfold, remains understudied.

To tackle this key research question, we developed the Integrated Social Ecological rEsilient lanD Systems (InSEEDS) model (v0.2), an agent-based dynamic global vegetation model (AB-DGVM). InSEEDS is a World-Earth Model that comprises an environmental and a socio-cultural component. Through a bidirectional tight coupling facilitated by the novel copan:lpjml framework (Breier et al., in prep), the Lund Potsdam Jena managed Land (LPJmL) is integrated as the Environmental component of InSEEDS. The model’s socio-cultural component is represented by an agent-based model (ABM) to simulate farmers’ management decisions for or against conservation tillage, which is a RA practice. 

This presentation introduces the InSEEDS model and describes its model set-up and design. It delineates the ABM structure, focussing on the farmers’ decision-making process, as well as the different farmer types. It lays out results regarding the co-evolutionary interactions in which the adoption and spreading dynamics are rooted in, the role of social and ecological heterogeneity in the model, as well as the interplay of factors in the agent decision-making.

The development and application process of the InSEEDS model points to future research directions for our model development team. In future model versions, the integration of qualitative and quantitative empirical knowledge for the ABM structure as well as parameterization could be an asset in comparison to the purely theoretical approach v0.2 follows. To capture more broad management options in the ecological component, the implementation as well as bundling of further RA practices like cover crops and agroforestry is planned (Breier et al., in prep). Furthermore, loop learning processes, more diverse agent types, non-local farmer networks, and higher levels of social organisation are envisioned as extensions of the farmer ABM component (Schwarz et al., in prep; Prawitz et al., in prep).