Abstract:
Agriculture accounts for over 10% of direct greenhouse gas emissions, with synthetic fertilizers in the US being a significant contributor. Digital agriculture offers promising solutions to enhance resource efficiency and mitigate the environmental impacts of production agriculture. This presentation will explore how the Digital Twin technology and Multi-Model Ensemble are being applied to estimate the net climate benefits of agricultural management practices at scale.
Bio:
Dr. Bruno Basso is a Hannah University Distinguished Professor at Michigan State University.
His research focuses on the long-term sustainability of agricultural systems, modeling of water, carbon, and nutrients fluxes across agricultural landscapes under current and future climates.
He is a Fellow of the American Association for the Advancement of Science (AAAS), a Fellow of the American Society of Agronomy, and a Fellow of the Soil Science Society of America. He received the 2021 Morgan Stanley Sustainable Solutions Prize and the 2025 Tech Transfer Achievement Award by MSU.
He is the co-founder of CIBO Technologies.
He is a member of the Board of Agriculture and Natural Resources of the US National Academies of Sciences, and Office of Science of the US Department of Energy.
He has published more than 250 scientific papers. He received his PhD from Michigan State University.
Summary:
Challenges:
Grow nutritious food with fewer resources in changing climate
Protect environment: soil, water, air quality, water quality, biodiversity
Reach negative GHG emissions
Reliable revenue stream for farmers to incentivize positive changes
Threats: urbanization, deforestation/desertification, fires/land degradation, soil erosion
Breakthrough in Agriculture: mechanization, fertilizers/agrochemicals, breeding/biotech, data science&modeling
Present: data science, circularity and sustainability
Focus: digital twins for farms for scaling solutions
Sensing, yield stability, multi-model ensembles, carbon intensity
Yield stability maps
Map yield of different pixels within the same field
Some subsets of fields are consistently more/less productive, others are unstable
Water availability
Opportunity for targeted treatment: more efficient use of inputs
Can relate other covariates (e.g. temperature) to yield, per pixel
Used on-ground data to train satellite model of yield stability; applied to 80M acres in US Midwest
48% stable high productivity, 25% stable low, 27% unstable
Locations of stable low regions is consistent across years, even when fertilizer is applied
Collected data on fertilizer application at fine resolution, related this to yield
Data shows that application of more fertilizer in stable low zones does not help yield
Precision Conservation
Profitability map from corn
-> Variable fertilizer application rate based on yield
-> Changes in nature impacts (carbon, biodiversity, etc)
Together->Profitability from both yield and carbon/biodiversity payments
Scope 3 GHG emissions
Climate impacts are being increasingly disclosed by companies
Challenge: measurement and modeling of Soil Organic Carbon (SOC)
Measurements: lab procedures, spatial variability, bulk density, remote sensing, spectroscopy
Critical to account for soil bulk density when measuring SOC
Deeper soils are more dense
Models: Process-based, ML+Process-based
Measured SOC across Midwest
Multi-model ensemble to model Midwest growing practices
Relating tillage, crop rotation, fertilizer, cover cropping
Critical to use dynamic baselines when analyzing interventions
Compare impact of climate smart in 10 years to conventional in 10 years, not conventional in today’s state
E.g. continuation of conventional may keep losing SOC
Models show that no-till and cover crops emit a little more GHGs but also sequester a lot more carbon in the soil, so the net impact is net-negative CO2