EAGER: Computational Agroecology: A Systems Approach

Recent work on network verification has had to deal simultaneously with the heterogeneity of large networks of hardware and software elements that interact in complex ways and with scaling verification and synthesis for practical use in network planning and management. These challenges mirror the challenges in applying computing coherently to new challenges in agriculture.

The last eighty years have seen a dramatic narrowing to just 7 crops that produce 80% of global calories, a situation that is insecure given increasing demands and changing pressures. This project will introduce a new computational framework extending and expanding recent networking research techniques to unify the disparate approaches to Digital Agriculture, with an aim to yield both greater productivity and greater security in the food system while advancing the state of the art in modeling agroecosystems computationally at large scale.

As a result of this large scale, we will develop new techniques for scaling network verification, especially in settings where only approximate data is available. The innovations developed in this research will be applicable back to large-scale network verification, planning, and provisioning.

Specifically, this project will introduce a computational framework for Digital Agriculture that subsumes both precision agriculture and agroecology. This computational framework will proceed to root the analysis, simulation, and understanding of agroecosystems using a network-verification-based space-time state-space representation of the infinite possible configurations of a piece of land and the biogeochemical elements on it.

This approach enables consideration of agroecological designs and systems of management, including complex mixtures of crops and cropping systems, that are seldom considered in conventional approaches; it simultaneously enables rigorous analysis of formerly inscrutable agroecological methods. This new framework will provide essential guidance for the critical changes facing vast human-managed lands.

This project will consist of a conceptual state-space framework called Agroecological Transition Functions and a practical software systems framework called Computational Agroecology. Beyond simply advancing agroecological understanding, this project will advance networked systems research by exploring state-space exploration, such as in network verification, at much larger scale than before, and by doing so considering the application of new types of networked systems of sensing and actuation in a complex physical environment.

This framework will be instantiated through new abstractions for programming cyberinfrastructure to explore new engineered technologies in sensing and actuation, including human practices and technologies that do not yet have physical instantiations. The outputs of this research will apply back to core areas of networked systems research; specifically, this work will enable improved scaling of network verification and synthesis through improvements in approximation and aliasing via the state-space framework that will be developed.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.