Select Page

 

SMARTFARM Project

 

Project Overview

icon-natureicon-solar

The SMARTFARM Project fits into the iSEE research themes of Secure & Sustainable Agriculture, Climate Solutions, and Energy Transitions.

 

In Lead Investigator Kaiyu Guan’s words:

“Our DOE-funded project team will develop a precise system for measuring greenhouse gas emissions from commercial bioenergy crops — essentially new technology for managing bioenergy crops, improving yield, reducing over fertilization, and designing new tools for ‘Smart Farms.’

“We will establish the Midwest Bioenergy Crop Landscape Laboratory (MBC Lab) to monitor emissions on fields in Champaign County, and the vast amount of data collected will be publicly available and could someday lead to financial rewards for farmers who reduce emissions through sustainable crop management.

“Our team will develop protocols for data processing and storage, create an online portal for growers and other researchers to access, and build the cyberinfrastructure for real-time data visualization. The researchers will also share findings regularly with the broader farmer communities in Illinois.

“Our MBC Lab is the perfect testbed for measuring emissions from biofuel production, and the platform we will develop will provide data and technology that can be applied to existing markets and broader production agriculture for ‘smart farming’ and environmental sustainability.”


Project News

SMARTFARM researchers were part of a multi-institutional team that has significantly improved the performance of numerical predictions for agricultural nitrous oxide emissions using novel modeling that combines artificial intelligence and process-based knowledge.

Researchers have developed a first-of-its-kind knowledge-guided machine learning model for agroecosystems called KGML-ag, which includes less obvious variables such as soil water content, oxygen level, and soil nitrate content related to nitrous oxide production and emission. Credit: iStock

According to a release from the University of Minnesota — home of the corresponding author and Digital Agriculture Group lead Zhenong Jin — the team developed a first-of-its-kind knowledge-guided machine learning model for agroecosystem, called KGML-ag, which is 1,000 times faster than current solutions and also significantly improves the modeling accuracy of greenhouse gas emissions from agriculture.

The research was recently published in Geoscientific Model Development. Researchers involved were from Minnesota, Illinois, Lawrence Berkeley National Laboratory, and the University of Pittsburgh. According to the Minnesota news release, KGML-ag was constructed by a special procedure that incorporates the knowledge learned from an advanced agroecosystem computational model, called ecosys, to design and train a machine learning model. In small, real-world observations, the KGML-ag turns out to be much more accurate than either ecosys or pure machine learning models.

“This is revolutionary work that brings together the best of observational data, process-based models, and machine learning by integrating them together,” said SMARTFARM Project Director Kaiyu Guan, a coauthor of the study, Founding Director of the Agroecosystem Sustainability Center (ASC) and Blue Waters Associate Professor of Natural Resources & Environmental Sciences (NRES) at Illinois. 

Jin, Assistant Professor of Agroecosystem Modeling at Minnesota, and Vipin Kumar, Professor and Head of Computer Science and Engineering at Minnesota, are co-authors from the SMARTFARM project. ASC and SMARTFARM team member Bin Peng and ASC team member Wang Zhou, both from NRES at Illinois, also were among the article’s 16 authors. 

Full iSEE news release >>>

Full University of Minnesota article >>>

The paper in Geoscientific Model Development >>>

There is much more carbon stored in Earth’s soil than in its atmosphere. A significant portion of this soil carbon is in organic form (carbon bound to carbon), called soil organic carbon (SOC). Notably, unlike the inorganic carbon in soils, the amount of SOC, and how quickly it is built up or lost, can be influenced by humans. Since its advent about 10,000 years ago, agriculture has caused a significant amount of SOC to be released into the atmosphere as carbon dioxide, contributing to climate change. 

Quantifying the amount of SOC in agricultural fields is therefore essential for monitoring the carbon cycle and developing sustainable management practices that minimize carbon emissions and sequester carbon from the atmosphere to the soil to reduce or reverse the climate effects of agriculture.

The traditional and most reliable way to quantify SOC is by soil sampling, with analyses in the lab (“wet chemical” measurement). But which locations in the field should be sampled? And how many samples should be taken for an accurate estimate? Each additional soil core adds significant labor and expense — and uncertainties in how to optimize sampling can lead to substantial extra costs.

In a new publication in Geoderma from the U.S. Department of Energy’s (DOE) SMARTFARM Project, researchers evaluated strategies for estimating SOC. Their goal was to develop an estimation strategy that maximizes accuracy while minimizing the number of soil cores sampled. 

View the full news release >>>

New machine learning algorithms could replace in-field soil sampling to accurately estimate soil organic carbon.

Just how much carbon is in the soil? That’s a tough question to answer at large spatial scales, but understanding soil organic carbon at regional, national, or global scales could help scientists predict overall soil health, crop productivity, and even worldwide carbon cycles.

Classically, researchers collect soil samples in the field and haul them back to the lab, where they analyze the material to determine its makeup. But that’s time- and labor-intensive, costly, and only provides insights on specific locations. 

In a recent study partially funded by the U.S. Department of Energy for the SMARTFARM project, University of Illinois researchers show new machine-learning methods based on laboratory soil hyperspectral data could supply equally accurate estimates of soil organic carbon. Their study provides a foundation to use airborne and satellite hyperspectral sensing to monitor surface soil organic carbon across large areas.

The lead study author is SMARTFARM’s Sheng Wang, Research Assistant Professor in Natural Resources & Environmental Sciences (NRES) and the Agroecosystem Sustainability Center (ASC). Other authors from SMARTFARM include Principal Investigator Kaiyu Guan, ASC Founding Director and NRES Associate Professor, D.K. Lee, Professor of Crop Sciences, and Jian Peng, Assistant Professor of Computer Science.

Read the College of ACES article by Lauren Quinn here >>>

Read the full paper in Remote Sensing of the Environment >>>

In a first-of-its-kind study, members of the SMARTFARM team and other University of Illinois researchers put hyperspectral sensors on planes to quickly and accurately detect nitrogen status and photosynthetic capacity in corn.

“Field nitrogen measurements are very time- and labor-consuming, but the airplane hyperspectral sensing technique allows us to scan the fields very fast, at a few seconds per acre. It also provides much higher spectral and spatial resolution than similar studies using satellite imagery,” says Sheng Wang, Research Assistant Professor in the Agroecosystem Sustainability Center (ASC) and the Department of Natural Resources and Environmental Sciences (NRES) at the U of I. Wang is lead author on the study; SMARTFARM Principal Investigator and ASC Director Kaiyu Guan and Research Scientist Chongya Jiang are among other co-authors.

The plane, fitted with a top-of-the-line sensor capable of detecting wavelengths in the visible and near infrared spectrum (400-2400 nanometers), flew over an experimental field in Illinois three times during the 2019 growing season. The researchers also took in-field leaf and canopy measurements as ground-truth data for comparison with sensor data.

The flights detected leaf and canopy nitrogen characteristics, including several related to photosynthetic capacity and grain yield, with up to 85% accuracy — “close to ground-truth quality,” Guan said.

Read the full article by the College of ACES’ Lauren Quinn >>>

Check out the new publication in the International Journal of Applied Earth Observation and Geoinformation >>>

Cornfield tissue sampling at the V6 (initiation of the uppermost ear and tassel) growth stage.

FROM MBC LAB WORK:

Guan reports that initial eddy covariance towers were set up in Bondville (October 2020) and two other sites (March 2021) — and that these three flux systems for CO2 and H2O based on Licor smartflux system have been functioning continuously and smoothly, with the exception of a minor hardware issue at one site and brief periods of seeding and fertilizer application in all the sites during May and June in 2021.

“We carry out routine inspection both in site and remotely, as well as system maintenance, including sensor/mirror cleaning, weed control, and other site management practices along the growing season,” he said.

Besides in-situ flux measurement based on the eddy covariance system, the team also carries out routine measurement on LAI and plant canopy heights in all sites.

Plant tissue sampling has been completed in all three locations at the V6 and VT (tassel) corn growth stages. Greenhouse gas sampling collection changed from weekly to biweekly since the emission decreased. Water and plant samples in process.

FROM SYMFONI WORK:

Among the many accomplishments during the past year, the team reports the following from the U of I sites:

  • Team members are developing two separate approaches to provide soil moisture information at high spatial resolution: to downscale the coarse-resolution passive radiometer data with optical and thermal signals; and to retrieve soil moisture by synergetic use of sentinel-1 SAR and optical remote sensing data. The team collected in-situ surface soil moisture and roughness measurements across multiple fields near the ARPA-E sites and the data will be used to validate the high-resolution soil moisture estimations.
  • To collect ground truth of leaf area index, researchers have built a ground camera network in the study area in Champaign County. The cameras automatically take downward viewing images from the top of canopy and upward viewing images from the bottom of canopy. These images are automatically analyzed to quantify leaf area index using computer vision technique and radiative transfer theory. This data  will be used to calibrate remote sensing models for the estimation of leaf area index for individual fields across the study area.
  • Team members have collected rich site-level historical observations at seven flux tower sites in the U.S. Midwest, which include biomass for different plant organs, leaf area index, net ecosystem exchange (NEE), and partitioned ecosystem respiration (Reco) and gross primary productivity (GPP) etc. We have used this data to calibrate the Ecosys model at the site scale.
  • Researchers generated synthetic data for corn-soybean rotations from 2001- to ’18 for 99 randomly selected sites by the Ecosys model. Hourly meteorological inputs (net radiation, air temperature, precipitation, relative humidity, and wind speed) were retrieved, and layerwise soil properties were derived to conduct the Ecosys simulations. Twenty different N fertilizer rates ranging from 0 to 300 lb N/acre were applied in May for each simulation to include the impacts of the N fertilizer application. Variables including daily N2O fluxes, N fertilizer rate, over 100 intermediate variables, weather forcings (converted to daily), and static soil properties were collected, and then processed into the matrix to input to the DL models

NEW PUBLICATION!

Computational models that track carbon as it cycles through an agroecosystem have massive untapped potential to advance the field of precision agriculture, increasing crop yields and informing sustainable farming practices.

“Although modeling the carbon cycle in agroecosystems has been done before, our work represents the most comprehensive integration of models and observations, as well as rigorous validation that includes rich measurements from both field and regional scales. The modeling performance of our solution (published this month in Agriculture and Forest Meteorology) far surpasses prior studies,” said SMARTFARM Lead Kaiyu Guan, an Associate Professor of Natural Resources & Environmental Sciences at the University of Illinois Urbana-Champaign. Guan is also a Blue Waters Professor at the National Center for Supercomputing Applications (NCSA) and Founding Director of the Agroecosystem Sustainability Center created by the College of Agricultural, Consumer, and Environmental Sciences and iSEE.

Read a full news release on this publication >>>

A postdoctoral research position is available in the Institute for Sustainability, Energy, and Environment at the University of Illinois Urbana-Champaign as part of the SMARTFARM project funded by the U.S. Department of Energy. A major goal of SYMFONI (System of Systems Solutions for Commercial Field-Level Quantification of Soil Organic Carbon and Nitrous Oxide Emission for Scalable Applications) is to understand the field scale mechanisms driving carbon storage and loss and gaseous nitrogen fluxes from commercial agricultural fields under different management practices.

The postdoc will focus on making high-resolution measurements of soil carbon dioxide and nitrous oxide fluxes with automated gas exchange equipment, as well as soil driving variables (e.g. pH, soil bulk density, soil organic matter and carbon, nutrients, etc). Experience with instrumentation and management of large data sets are desirable for this position. A willingness to learn techniques for spatial analyses also would be beneficial. The successful candidate will have experience conducting field research and should be comfortable working in a highly collaborative environment where they will coordinate research activities with a large team, supervise technicians and undergraduates, and write manuscripts. Field work will be conducted on commercial farms in Champaign County, Illinois.

The postdoc will work primarily under the guidance of Co-PIs Evan DeLucia, Wendy Yang, and Carl Bernacchi, and will be expected to collaborate with other ecosystem ecologists, ecosystem modelers, ecohydrologists, and others in the SYMFONI project.

Required Qualifications:

  • A Ph.D. or the equivalent in ecology, biogeochemistry, soil science, or related field
  • Experience with laboratory and field work
  • Strong English writing and oral communication skills
  • Strong organizational skills
  • Ability to work in a collaborative environment
  • Ability to travel frequently to local research sites
  • A valid driver’s license

All candidates must have received a Ph.D. in a relevant field within the past five years. The position is available for two years, with possible extension further; however, annual renewal is dependent on funding and progress made by the individual. This position includes a competitive salary and full benefits.

Application review will begin June 15, 2021, and will continue until the position is filled. The start date is flexible, but ideally the start would be on or before Dec. 1, 2021. Applications should include a brief cover letter, curriculum vitae, and the names and contact information for three references. Please send your application via email to Anya Knecht, the SYMFONI Research Coordinator, at knecht2@illinois.edu.

For further information about the position, please contact DeLucia at delucia@illinois.edu, Yang at yangw@illinois.edu, or Bernacchi at bernacch@illinois.edu.

Illinois is an Affirmative Action /Equal Opportunity Employer and welcomes individuals with diverse backgrounds, experiences, and ideas who embrace and value diversity and inclusivity (www.inclusiveillinois.illinois.edu).

The University of Illinois conducts criminal background checks on all job candidates upon acceptance of a contingent offer.

On Sept. 8, 2020, the University of Illinois announced that the Smart Farms team led by Kaiyu Guan was awarded $4.5 million from the U.S. Department of Energy’s Advanced Research Projects Agency-Energy (ARPA-E) through its “Systems for Monitoring and Analytics for Renewable Transportation Fuels from Agricultural Resources and Management” (SMARTFARM) program. 

The funding will be used to calculate farm-scale carbon credits, allowing individual farmers to understand the value of their land and practices toward carbon trading markets. Funding for the newest phase of the project, named “SYMFONI,” allows accurate and rapid field-level quantification of carbon intensity for every individual field across the U.S. and can be seamlessly scaled up to the global scale. This is made possible through the integration of field-based observations with satellite and aerial hyperspectral data, physics-guided deep learning, mobile soil sensing, and supercomputing. Guan, Assistant Professor of Natural Resources and Environmental Sciences (NRES) and a Blue Waters Professor at the National Center for Supercomputing Applications (NCSA), says the system builds a generic framework that will flexibly integrate newer sensor technologies as they become available, meaning the output will continue to improve over time.

“We couldn’t be prouder of Kaiyu and his team’s efforts toward a more sustainable future for agriculture and the planet. Innovations from researchers within the College of Agricultural, Consumer and Environmental Sciences (ACES), along with our partners across campus, have been changing the world for 150 years, and with investments like this from ARPA-E, we’ll continue moving forward to a brighter future,” said German Bollero, Associate Dean for research in the College of ACES.

The Illinois team was the only group nationwide to be twice awarded funds in ARPA-E’s SMARTFARM program in its two funding phases. The same team leads a Phase 1 project ($3.3 million) to collect gold-standard carbon emission data at the farm scale and build the foundation for testing Phase 2 technologies. The Illinois-led project also represents the largest portion of SMARTFARM funding, receiving 27% of the total distributed in the current round of funding.

The project team includes nine members across Illinois’ Urbana-Champaign campus. For this round of funding, Jian Peng from Grainger College of Engineering was added — as were Zhenong Jin and Vipin Kumar from the University of Minnesota, Kang Sun from the University of Buffalo, and Jinyun Tang from Lawrence Berkeley National Lab.

Read the full news announcement >>>

The Institute for Sustainability, Energy, and Environment (iSEE) was instrumental in helping a University of Illinois team land a $3.3 million U.S. Department of Energy grant.

The U.S. Department of Energy has awarded a $3.3 million grant to a multidisciplinary research team at the University of Illinois at Urbana-Champaign to develop a precise system for measuring greenhouse gas emissions from commercial bioenergy crops grown in central Illinois.

The three-year project through the Institute for Sustainability, Energy, and Environment (iSEE) is expected to reduce emissions associated with ethanol and other biofuels by enabling new technology for managing bioenergy crops, improving yield, reducing overfertilization, and designing new tools for “smart farms.” The vast data collected will be publicly available and could someday lead to financial rewards for farmers who reduce emissions through sustainable crop management.

Led by Kaiyu Guan, an Assistant Professor of Natural Resources and Environmental Sciences (NRES), the team will establish the Midwest Bioenergy Crop Landscape Laboratory (MBC-Lab) to monitor emissions on three 85-acre maize and soybean fields in Champaign County.

Read the full news announcement >>>


The Team

Principal Investigator (PI) and co-PIs

  • MBC Lab and SYMFONI: Kaiyu Guan (PI), Associate Professor of Natural Resources & Environmental Sciences, University of Illinois Urbana-Champaign (right)
    Departmental page >>>
    Lab page >>>
  • MBC Lab and SYMFONI: Carl Bernacchi, Plant Physiologist, U.S. Department of Agriculture Agricultural Research Service
    USDA ARS page >>>
    Lab page >>>
  • MBC Lab and SYMFONI: Evan H. DeLucia, Arends Professor Emeritus of Plant Biology, University of Illinois Urbana-Champaign
    Departmental page >>>
    Lab page >>>
  • MBC Lab: Jeremy Guest, Associate Professor of Civil & Environmental Engineering, University of Illinois Urbana-Champaign
    Departmental page >>>
    Lab page >>>
  • SYMFONI: Zhenong Jin, Assistant Professor of Agroecosystem Modeling, University of Minnesota
    Departmental page >>>
    Lab page >>>
  • MBC Lab and SYMFONI: D.K. Lee, Professor of Crop Sciences, University of Illinois Urbana-Champaign
    Departmental page >>>
    Lab page >>>
  • MBC Lab: Jong Lee, Principal Research Scientist at National Center for Supercomputing Applications, University of Illinois Urbana-Champaign
    NCSA page >>>
  • SYMFONI: Jian Peng, Assistant Professor of Computer Science, University of Illinois Urbana-Champaign
    Departmental page >>>
  • MBC Lab and SYMFONI: Wendy Yang, Associate Professor of Plant Biology, University of Illinois Urbana-Champaign
    Departmental page >>>
    Lab page >>>

 

Operating Team (Faculty, Postdocs, Technicians, Students)

  • MBC Lab: Chunhwa Jang, Postdoc in Crop Sciences, University of Illinois Urbana-Champaign
  • MBC Lab and SYMFONI: Chongya Jiang, Research Scientist at the Institute for Sustainability, Energy, and Environment (iSEE), University of Illinois Urbana-Champaign
  • SYMFONI: Vipin Kumar, William Norris Professor and Head of Computer Science and Engineering, University of Minnesota
    Departmental page >>>
  • SYMFONI: Kang Sun, Assistant Professor of Civil, Structural and Environmental Engineering, University at Buffalo
    Departmental page >>>
    Lab page >>>
  • MBC Lab: Yuchen Liu, Postdoc in Natural Resources & Environmental Sciences, University of Illinois Urbana-Champaign
  • MBC Lab: Xiangmin “Sam” Sun, Postdoc in Plant Biology, University of Illinois Urbana-Champaign
    Bernacchi Lab page >>>
    Read a profile on Sam and his SMARTFARM work >>>
  • SYMFONI: Jinyung Tang, Research Scientist, Lawrence Berkeley National Laboratory
    Departmental page >>>
  • MBC Lab and SYMFONI: Sheng Wang, Research Assistant Professor in Natural Resources & Environmental Sciences, University of Illinois Urbana-Champaign
    Read a Carl R. Woese Institute for Genomic Biology profile on Sheng >>>
  • MBC Lab and SYMFONI: Bin Peng, Research Scientist in Natural Resources & Environmental Sciences, University of Illinois Urbana-Champaign
  • SYMFONI: Jacob Forbes, Research Technician, iSEE
  • MBC Lab: Leslie Stoecker, Data Manager, Center for Advanced Bioenergy and Bioproducts Innovation, University of Illinois Urbana-Champaign
  • SYMFONI: Chantelle Lonsdale, Ph.D. Candidate, Illinois

 

Project Manager

  • MBC Lab and SYMFONI: Anya Knecht, Institute for Sustainability, Energy, and Environment (iSEE), University of Illinois Urbana-Champaign

 

Collaborator

  • MBC Lab: Tilden Meyers, National Oceanic and Atmospheric Administration Atmospheric Turbulence and Diffusion Division, Oak Ridge, Tenn.

Publications

(iSEE project members’ names in bold):

  • “KGML-ag: A Modeling Framework of Knowledge-Guided Machine Learning to Simulate Agroecosystems: A Case Study of Estimating N20 Emission Using Data from Mesocosm Experiments, Liu, L., Xu, S., Tang, J., Guan, K., Griffis, T.J., Erickson, M.D., Frie, A.L., Jia, X., Kim, T., Miller, L.T., Peng, B., Wu, S., Yang, Y., Zhou, W., Kumar, V., Jin, Z. Geoscientific Model Development (April 2022); read the University of Minnesota news release >>> 
  • “How to Estimate Soil Organic Carbon Stocks of Agricultural Fields? Perspectives Using Ex-Ante Evaluation,” Potash, E., Guan, K., Margenot, A., Lee, D.K., DeLucia, E.H., Wang, S., Jang, C. Geoderma (April 2022); read the iSEE news release >>> 
  • “Using Soil Library Hyperspectral Reflectance and Machine Learning to Predict Soil Organic Carbon: Assessing Potential of Airborne and Spaceborne Optical Soil Sensing,” Wang, S., Guan, K., Zhang, C., Lee, D.K., Margenot, A.J., Ge, Y., Peng, J., Zhou, W., Zhou, Q., Huang, Y. Remote Sensing of Environment (March 2022); read the ACES news release >>>
  • “Airborne Hyperspectral Imaging of Nitrogen Deficiency on Crop Traits and Yield of Maize by Machine Learning and Radiative Transfer Modeling.” Wang, S., Guan, K., Wang, Z., Ainsworth, E.A., Zheng, T., Townsend, P.A., Liu, N., Nafziger, E., Masters, M.D., Li, K., Wu, G., Jiang, C. International Journal of Applied Earth Observation and Geoinformation (December 2021); read the ACES news release >>>
  • “Quantifying Carbon Budget, Crop Yields, and their Responses to Environmental Variability Using the Ecosys Model for U.S. Midwestern Agroecosystems.” Zhou, W., Guan, K., Peng, B., Tang, J., Jin, Z., Jiang, C., Grant, R. Mezbahuddin, S. Agricultural and Forest Meteorology (July 2021); read the iSEE news release >>>
  • “Identifying Nitrogen Loss Hotspots and Mitigation Potential in the U.S. Corn Belt with a Metamodeling Approach.” Kim, T., Jin, Z., Smith, T., Liu, L., Yang, Y., Yang, Y., Peng, B., Phillips, K., Guan, K., Hunter, L.C. Environmental Research Letters (July 2021)
X