Satellite Technologies Aid Localized Planning

View your grain supply chain in higher definition
By Steffen Mueller and Ken Copenhaver | September 11, 2012

Recent severe weather events such as last spring’s flooding and ponding in fields and this summer’s heat stress and drought pose a challenge to grain buyers at ethanol plants. The geographic diversity of the U.S. Corn Belt provides a natural hedge against catastrophic national crop failures, putting the risk to corn supply chains squarely at the subregional level. Satellite imagery algorithms, originally developed by the University of Illinois at Chicago to refute overstated land use change claims set forth by many academics and regulators against biofuels production, are now being used in a novel way to protect corn supply chains for ethanol plants.

In 2007, many academics and regulators produced analyses concluding that then-emerging biofuels policies, including the current renewable fuel standard, would result in conversion of large amounts of native lands to crop land. Satellite imagery was used to support these claims. At that time, two of us from the University of Illinois at Chicago disagreed and published a string of reports in sensible support of biofuels production (Mueller and Copenhaver, 2009a). These reports laid out the proper use of satellite algorithms and data-vetting routines to accurately determine land use and land use change (Mueller and Copenhaver, 2009b; Mueller, Copenhaver, Begert, 2012).  What originally helped assuage the tense land use policy debate has now become a capable tool to help ethanol plants assess real-time corn availability and manage supply chain risk.

In the past, grain buyers have watched and relied heavily on USDA reports that are generated throughout the growing season including the weekly Crop Progress Reports, the Preliminary Acreage Report (usually released end of June) and the USDA Crop Report (usually released the second week of August). While these reports provide helpful benchmarks, they themselves influence market price movements. This means that by the time the reports are released, it is challenging to adjust risk management strategies. Also, the spatial resolution of the reports focuses on the larger interregional level (the August crop report releases only state-wide data) whereas ethanol plants and grain buyers benefit from county and even subcounty-level data. Since USDA relies heavily on farmer survey data (combined with some satellite imagery analysis), county- level acreage and yield data is only released in the spring following the growing season of interest. Also, the fixed release schedule for some of the key reports set by USDA months in advance may render some of the information less meaningful. For example, in an early planting year like this one, releasing preliminary yield estimates close to mid-August trails the emerging knowledge in the fields.

The research products developed at UIC have shown that satellite imagery can fill the spatial and temporal information void and provide useful and reliable information for grain buyers such as ethanol plants. Roughly 200,000 acres of corn are needed in a tight radius to supply a 100 MMgy plant (although the distillers grains feed coproducts offset half of that acreage). As shown in the 2008 National Corn Ethanol Survey (Mueller, 2010), ethanol plants, on average, source grain within a 50-mile area from their plants but the individual corn supply radii vary widely. In fact, many ethanol plants draw corn from skewed polygon areas rather than circular ones. Besides transportation logistics, the reasons for that variation are different cropping rotations (amount of double cropping in the vicinity), prevailing yields in the area, disease pressure, and, influencing all of these: local weather patterns. Reliable in-season, subregional data allows ethanol plants to determine early the proper supply radius (or polygon structure) and those clusters of growers or grain elevators that are affected by above- or below-average yields. The weather patterns of the past two years (floodings of 2011 and the heat pattern of 2012) accentuate the trend to move towards subregional data information sources.

In comes the LandViewer geospatial platform developed by UIC. LandViewer allows a very localized view of an ethanol plant’s corn supply area. It calls on proprietary software algorithms to rapidly process new satellite imagery and weather station data for comparison with databases of vegetation, climate and planting history. Three types of satellite imagery at different resolutions are used to calibrate the yield and acreage models. Additional ground verification (ground-truthing) is performed during scouting trips in early June by cataloguing the coordinates of corn fields for input to the satellite acreage model and in late July during scouting trips by agronomists for input to the yield model. The highest resolution satellite data is also used to work closely with key growers delivering to the plants. Then, in a series of meetings with the ethanol plants and their key growers, the LandViewer team provides several information products: near-infrared field conditions (Figure 1), vegetation vigor in corn draw area (Figure 2), acreage estimates (Figure 3), yield estimates (Figure 4) and localized weather conditions (Figure 5).

The meetings with the ethanol plants are strategically scheduled throughout the growing season to optimize the plants’ risk management strategies. Currently, LandViewer technologies are used by three independent ethanol plant operators operating six plants across Illinois and Iowa. The research effort, now in its second season, provides valuable and timely information to the plants. The results from the first season showed accurate yield and acreage prediction prior to USDA report releases. Rick Schwarck, CEO of Absolute Energy in Iowa states: “The information provided to us by the UIC LandViewer research team is very valuable since it constitutes an independent information data source specifically tailored to our corn supply area and our risk management schedule.”

The LandViewer team has recently completed substantial automation and programming efforts and is now in the process of rolling out the technology on a national scale. The yield and acreage models and the data display are protected by a provisional patent held by UIC. The information hub for the LandViewer Platforms is provided at http://www.landviewer.uic.edu/. Talks are in progress to commercialize the technology.

Outside of this LandViewer effort, the research team has a long tradition of support from the agricultural community including recently funded efforts by Monsanto, the Illinois Corn Marketing Board and the Iowa Corn Promotions Board.

Authors: Steffen Mueller
Principal Economist
University of Illinois at Chicago Energy Resources Center
muellers@uic.edu
(312) 316-3498

Ken Copenhaver
Senior Research Engineer
University of Illinois at Chicago Energy Resources Center
kcopenha@uic.edu
(217) 377-0071



References:
Mueller, S. and Ken Copenhaver (2009a). Letter to the Honorable Lisa Jackson, U.S. EPA, dated September 21, 2009, filed in Docket ID No. EPA-HQ-OAR-2005-0161 as part of the Renewable Fuel Standard 2 Program.

Mueller, Steffen and Copenhaver, Kenneth “Use of Remote Sensing to Measure Land Use Change from Biofuel Production, published in “The bulletin of the Program in Arms Control, Disarmament, and International Security”; University of Illinois at Urbana–Champaign, Volume XVII / No. 2 / Summer 2009b.

Mueller, S. “2008 National dry mill corn ethanol survey”; Biotechnol Lett DOI 10.1007/s10529-010-0296-7, May 15, 2010.

Mueller, S., Ken Copenhaver, and Dan Begert “An Assessment of Available Lands for Biofuels Production in the United States Using USDA Cropland Data Layers”, Journal of Agricultural Extension and Rural Development, 2012.

 

The claims and statements made in this article belong exclusively to the author(s) and do not necessarily reflect the views of Ethanol Producer Magazine or its advertisers. All questions pertaining to this article should be directed to the author(s).