USDA and NASA sign a data agreement to sharpen US acreage and yield estimates
USDA and NASA signed an MOU on 24 October 2026 to put satellite data and machine learning into US crop acreage, yield and disaster estimates.
The United States Department of Agriculture and NASA signed a memorandum of understanding in Washington on 24 October 2026, issued as release number 0142.26. Secretary of Agriculture Brooke L. Rollins and NASA Administrator Jared Isaacman signed it, and it took effect immediately. The agreement commits the two agencies to joint work on Earth observation, geospatial analytics and statistical methods for agricultural statistics. Named as possible areas of cooperation are agricultural statistics, remote sensing science, modelling, geospatial data engineering and machine learning, and USDA says the list is not exhaustive and can be extended as priorities change.
Three areas of cooperation are set out in the text. Mission and data alignment covers regular technical exchanges to align USDA statistical requirements with NASA Earth science missions and data products, including satellite observations, model outputs and derived geophysical variables. Data access and infrastructure covers scalable access patterns to NASA archives, cloud-native geospatial formats, application programming interfaces and the workflow orchestration that USDA production operations would need. Methodology and validation covers joint technical reviews of algorithms for crop condition indicators, yield models, planted and harvested area estimation and disaster impact assessment, including bias correction, uncertainty quantification and validation using statistically sound frames and ground data.
The memorandum sits under the USDA Data Modernization Plan that Rollins launched in September 2026, whose stated aims are to put farmers first, reduce the reporting burden on producers and improve the timeliness, accuracy and usefulness of agricultural data. USDA describes the two agencies as long-standing partners and points to Executive Order 14369, “Ensuring American Space Superiority”, issued in December 2025, as the trigger for a wider set of opportunities. The practical target named in the release is better acreage and yield estimation.
What the document does not contain is money or dates. USDA attached no dollar figure and no milestones beyond the immediate effective date, and the release describes mechanisms rather than deliverables. Rollins said: “American farmers are the best in the world and deserve any solution we can leverage, even when it is out of this world.” Isaacman said: “NASA develops some of the most advanced sensing, computing, and analytical capabilities in the world, and we should put those tools to work wherever they can deliver meaningful outcomes for the American people.” The two agencies also signed a memorandum of understanding on agricultural and Earth science cooperation in December 2020, so this is a widening of an existing relationship rather than a first step.
Why it matters for AgTech founders: the public baseline for crop intelligence is moving, and that resets what a satellite analytics product can charge for. USDA acreage and yield estimates are the reference that insurers, lenders and grain buyers price against, and if those numbers get sharper and more frequent on the back of NASA archives, a generic field-level yield dashboard loses most of its reason to exist. Two practical consequences. First, if the pitch is that you estimate area and yield from Sentinel and Landsat imagery, assume a free comparator of government quality in the largest market and move the product closer to the decision, which means variable-rate prescriptions, claims handling or contract triggers rather than the estimate itself. Second, read the data access section as a procurement signal: cloud-native formats, interfaces and workflow orchestration are what USDA says it needs, and agencies that write that sentence down tend to buy the plumbing rather than build all of it. For European founders the mirror question is whether Copernicus-based products will get the same institutional pull. So far the EU has the imagery without an equivalent statistical customer.
USDA and NASA sign a data agreement to sharpen US acreage and yield estimates
Signed in Washington on 24 October and effective immediately, the memorandum pulls NASA Earth observation data, geospatial engineering and machine learning into the statistics behind US acreage, yield and disaster estimates. No money is attached.
The United States Department of Agriculture and NASA signed a memorandum of understanding in Washington on 24 October 2026, issued as release number 0142.26. Secretary of Agriculture Brooke L. Rollins and NASA Administrator Jared Isaacman signed it, and it took effect immediately. The agreement commits the two agencies to joint work on Earth observation, geospatial analytics and statistical methods for agricultural statistics. Named as possible areas of cooperation are agricultural statistics, remote sensing science, modelling, geospatial data engineering and machine learning, and USDA says the list is not exhaustive and can be extended as priorities change.
Three areas of cooperation are set out in the text. Mission and data alignment covers regular technical exchanges to align USDA statistical requirements with NASA Earth science missions and data products, including satellite observations, model outputs and derived geophysical variables. Data access and infrastructure covers scalable access patterns to NASA archives, cloud-native geospatial formats, application programming interfaces and the workflow orchestration that USDA production operations would need. Methodology and validation covers joint technical reviews of algorithms for crop condition indicators, yield models, planted and harvested area estimation and disaster impact assessment, including bias correction, uncertainty quantification and validation using statistically sound frames and ground data.
The memorandum sits under the USDA Data Modernization Plan that Rollins launched in September 2026, whose stated aims are to put farmers first, reduce the reporting burden on producers and improve the timeliness, accuracy and usefulness of agricultural data. USDA describes the two agencies as long-standing partners and points to Executive Order 14369, “Ensuring American Space Superiority”, issued in December 2025, as the trigger for a wider set of opportunities. The practical target named in the release is better acreage and yield estimation.
What the document does not contain is money or dates. USDA attached no dollar figure and no milestones beyond the immediate effective date, and the release describes mechanisms rather than deliverables. Rollins said: “American farmers are the best in the world and deserve any solution we can leverage, even when it is out of this world.” Isaacman said: “NASA develops some of the most advanced sensing, computing, and analytical capabilities in the world, and we should put those tools to work wherever they can deliver meaningful outcomes for the American people.” The two agencies also signed a memorandum of understanding on agricultural and Earth science cooperation in December 2020, so this is a widening of an existing relationship rather than a first step.
Why it matters for AgTech founders: the public baseline for crop intelligence is moving, and that resets what a satellite analytics product can charge for. USDA acreage and yield estimates are the reference that insurers, lenders and grain buyers price against, and if those numbers get sharper and more frequent on the back of NASA archives, a generic field-level yield dashboard loses most of its reason to exist. Two practical consequences. First, if the pitch is that you estimate area and yield from Sentinel and Landsat imagery, assume a free comparator of government quality in the largest market and move the product closer to the decision, which means variable-rate prescriptions, claims handling or contract triggers rather than the estimate itself. Second, read the data access section as a procurement signal: cloud-native formats, interfaces and workflow orchestration are what USDA says it needs, and agencies that write that sentence down tend to buy the plumbing rather than build all of it. For European founders the mirror question is whether Copernicus-based products will get the same institutional pull. So far the EU has the imagery without an equivalent statistical customer.