Research

Societal Grand Challenge: Climate change is increasing the frequency and severity of storms, droughts, and forest fires. It is also raising precipitation variability, narrowing planting windows, and may lower yields as early as 2030. This will affect our ability to produce food, fiber, and fuel, whose demand is increasing due to population growth. There is a growing risk of major human, economic, and ecological toll on farmers, ranchers, forest managers, and consumers.

At the same time, agriculture and forestry have unrealized potential to reduce GreenHouse Gas (GHG) emissions and become a larger GHG sink to slow climate change. While our forests and farms have significant untapped potential for net GHG (e.g., atmospheric CO2) reduction, realizing this potential requires addressing competition between land for biomass production and ecosystem services, scientific and technological challenges of actionable (e.g., cheaper and more accurate) sequestered carbon verification, computational integration of the incentive structures of land managers that implement agriculture and forest management practices, and policy innovations to encourage large-scale equitable adoption of climate-smart practices.

Importance of specific aspects of this challenge we aspire to solve: We aspire to develop and implement AI tools and other recent breakthroughs in agriculture and forestry to improve understanding of trade-offs and feedback loops among crop productivity, human decision making, and GHG fluxes; develop cheaper and ore accurate GHG and biomass estimation methods; and create spatially explicit multiscale (field-to-market) decision support tools to explore tradeoffs between adaptation and mitigation. This will aid in realizing the untapped GHG-sink potential of agriculture and forestry.

The agriculture sector, for example, has the potential to reduce US GHG emissions by 12.4 percent. Forest management, afforestation and reforestation are also among the most promising negative emissions technologies (called interventions in this proposal), even though forests already are a major terrestrial carbon sink via natural regrowth after harvest, forest fire and other disturbances. And a functioning GHG market will sequester carbon and strengthen the rural economy. Overall, they will advance climate resilience, bio-productivity, and environmental benefits such as improved soil health and water quality while equitably enhancing equity and economy.

Timeliness of addressing the problem: Human, economic, ecological tolls and risk of irreversible tipping points are growing due to rising GHG concentrations and associated climate change. It is projected to lower yields as early as 2030. Longer we wait, the shorter time-window we have to avoid irreversible tipping points. Thus, there is an urgent need to counter climate-related threats, lower GHG emissions, and even become a bigger GHG sink. We must use all the tools at hand.

AI breakthroughs have enabled game-changing capabilities, including (in-situ and remote) sensors for GHG monitoring, data analytics and spatial-statistics to understand farm and forest variability, and early detection of disease in crops and forest. Compelling opportunities include next-generation AI (e.g., Knowledge-Guided Machine Learning, or KGML) to improve accuracy of GHG and soil organic matter estimates, improve understanding trade-offs between mitigation and adaptation practices as well as incentive structures, grow an AI workforce, and increase new rural opportunities in the emerging GHG market and related CSAF-tech sector.

Broader Impacts: This project is designed to ensure broad impacts by directly engaging with and addressing stakeholder needs regarding the development of user-inspired, AI-driven research on CSAF. We will leverage many existing collaborations between our team of AI and CSAF scientists and their curated datasets across our partner institutions, ensuring that the Institute can begin to immediately generate ground-breaking scientific insights and AI tools for CSAF. These will be used to train a new AgFoAI workforce, and serve as a nexus point for collaborations through education, communication, and outreach activities that reach undergraduate and graduate students, underrepresented minorities, diverse farmer, agribusiness, and commodity groups, and other important stakeholders working to speed the reduction of GHG emissions, increase adaptation, and sustainability increase food production to ensure nutrition security.

Climate-Smart Agriculture and Forestry (CSAF) Objectives: The focus of our research: combining the strengths of AI, modeling, and sensing techniques. The agriculture and forestry of the future must more sustainably increase productivity, enhance resilience (adaptation), reduce/remove GHGs (mitigation) where possible, and enhance food security and development goals. We believe this transformation can be facilitated through the combination of sensing, modeling, and AI technologies.

Sensing is a powerful tool for monitoring spatial and temporal variability in crop and animal stress, but has limited ability to penetrate soil, clouds, or animal hides. Process-based (PB) crop and forest simulation models are powerful tools for understanding how biophysical processes affect temporal patterns in growth, but a lack of fine-grained pervasive CSAF data limits scaling up PB models from plots to fields to watersheds and larger scales. AI is a powerful tool for describing and predicting complex agricultural and forestry data relationships, but is limited for modeling physical processes that may be evolving and interacting at multiple spatial and temporal scales. Addressing these challenges thus requires integrating the unique strengths of these technologies, creating vast synergies that drive innovation in CSAF and AI, as we illustrate through the following integrative vignettes.