USDA to Test Satellites and AI for Crop Estimates

1 hour ago

0 2 minutes read

By MBN Staff | MBN staff – Fri, 09/04/2026 – 13:12 Idioma Leer en Español DIA assistant 1.0x ✕

The US Department of Agriculture (USDA) will launch a pilot project using satellite imagery, geospatial tools, and crop models to improve its estimates of US crop acreage and yields, as the agency seeks to address growing concerns from farmers, grain traders, and economists over the reliability of its agricultural data.

Brooke Rollins, US Agriculture Secretary, announced the initiative at the Farm Progress Show in Boone, Iowa, following a month-long series of listening sessions with farmers and agricultural groups. The broader data modernization effort is intended to improve the accuracy and timeliness of USDA statistics while reducing the amount of information repeatedly requested from farmers.

“We will conduct a pilot to evaluate the use of improved satellite imagery, working with NASA and other federal government agencies across the country to figure out how we can make this reporting more accurate,” Rollins said during the conference.

Under the plan, the USDA will combine improved satellite imagery and other geospatial tools with crop models and farmer surveys to estimate planted acreage and yields. The department will also explore the use of AI and machine learning as part of the modernization process.

The initiative follows criticism of USDA agricultural statistics after deep staff cuts at the agency and unprecedented differences between initial and final estimates of the number of acres of corn planted and harvested by US farmers in 2025. The changes are also intended to strengthen transparency around the data collection process. The USDA said it would provide farmers with more information about its methodologies, survey response rates and data limitations.

Much of the criticism has focused on the National Agricultural Statistics Service (NASS), whose reports are closely monitored by agricultural markets. Its estimates can influence commodity prices, the prices farmers receive for their crops and federal agricultural programs. The impact of USDA data on commodity markets was highlighted after the agency released its final estimates in January, when already low grain prices fell more than 5% as growers faced difficulties generating profits.

Rollins has acknowledged the loss of confidence in USDA statistics and said the department needs to rebuild farmers’ trust in its data. Listening sessions launched at Minnesota Farmfest on Aug. 4 have provided farmers with an opportunity to raise concerns about the reliability of USDA statistics, staffing reductions and other policies affecting the farm economy. He plans to hold additional listening sessions through the fall, Harry Fones, Spokesman, USDA, told Reuters.

The modernization effort comes as the USDA seeks to balance the need for timely and accurate agricultural statistics with the burden of data collection on farmers. By incorporating satellite imagery, geospatial technology, crop modeling, AI, and machine learning alongside farmer surveys, the department aims to develop new methods for producing crop estimates while providing greater visibility into how the figures are generated.