Data sources

YieldVista is built on public satellite, agricultural, and weather datasets. We don't resell raw imagery or government data — we compute weekly indices, baselines, deviations, and yield outlooks from these sources, in the user interface. The list below covers what we ingest, who publishes it, and the terms under which we use it.

Satellite imagery

Copernicus Sentinel-2

Optical imagery from the European Space Agency's Sentinel-2A and Sentinel-2B satellites, accessed via the public STAC catalog. NDVI, EVI, and NDWI compute on the 10 m visible and near-infrared bands; NDMI uses the 20 m shortwave-infrared band alongside NIR. Two satellites in orbit since mid-2017, ~5-day revisit; our archive starts 2017. Aggregated to weekly composites.

Attribution: Contains modified Copernicus Sentinel data 2017–present. Provided by the European Space Agency under the Copernicus open data policy. License terms.

Crop masks

USDA Cropland Data Layer (CDL)

30 m annual raster of crop types, published by USDA NASS. We use CDL to isolate corn and soy pixels before computing aggregates so reads are crop-specific rather than landscape-average.

Attribution: USDA National Agricultural Statistics Service, Cropland Data Layer. Public domain. CDL homepage.

Yield ground truth

USDA NASS Quick Stats

County-level annual corn and soybean yields (bushels per acre) used as the supervisory target for our yield models and as the historical anchor for five-year baselines.

Attribution: USDA National Agricultural Statistics Service, Quick Stats database. Public domain. Quick Stats.

Weather + climate

IEMRE

Iowa Environmental Mesonet Reanalysis — daily gridded precipitation, temperature min/max, and derived variables on a ~14 km (1/8°) CONUS grid. Hosted by Iowa State University. We use IEMRE as the canonical weather source for county-level summaries and as model features.

Attribution: Iowa State University, Iowa Environmental Mesonet, IEMRE dataset. IEMRE.

Drought monitoring

U.S. Drought Monitor (USDM)

Weekly drought-classification map (D0–D4) produced by the National Drought Mitigation Center, USDA, and NOAA. We surface county-level USDM categories in commentary and overlay them where they explain a deviation in the satellite read.

Attribution: U.S. Drought Monitor, a partnership of the National Drought Mitigation Center at the University of Nebraska-Lincoln, the United States Department of Agriculture, and the National Oceanic and Atmospheric Administration. USDM.

Soil

SSURGO

USDA Natural Resources Conservation Service Soil Survey Geographic Database. We use SSURGO-derived attributes (available water capacity, drainage class, organic matter) as static features when training yield models.

Attribution: USDA Natural Resources Conservation Service, SSURGO database. Public domain. SSURGO.

Insurance / acreage

USDA RMA Summary of Business

County-level insured acres for corn and soybeans, published by USDA's Risk Management Agency. We use RMA SOB to acreage-weight county-level aggregations and as a sample weight for yield model training.

Attribution: USDA Risk Management Agency, Summary of Business. Public domain. RMA SOB.

Crop condition / progress

USDA NASS Crop Progress & Condition

Weekly state-level percentages of corn and soybean planting, silking, pod-set, maturity, and condition ratings (very poor through excellent). Published Mondays during the growing season. We compute pace and condition baselines from these reports.

Attribution: USDA National Agricultural Statistics Service, Crop Progress and Condition reports. Public domain. Crop Progress.