Estimating the health of any crop in the US is helped greatly by the large and high-quality database that the USDA maintains. Gro’s US soybean yield-modeling process ended up looking broadly similar to that of our US corn model. In both models, a 50-year linear trend yield estimate gave us the best results, and NDVI proved the best predictor of deviation from trend yield. However, due to the significant geographic range of the soybean crop all the way down the Mississippi River, soil diversity was more important than it was for corn. Soil moisture variables also played a bigger role, supplementing the greenness sensors that detect soybean health a little less definitively than corn health. In a repeat of the corn model experience, we saw the largest average errors in our estimates in counties that produced fewer soybeans on the fringes of the belt. Some variables selected by our experts include:

  • Normalized difference vegetation index (NDVI)
  • Land surface temperature (LST)
  • ET anomaly (monthly)
  • Rainfall data (TRMM)
  • SMOS
  • Crop calendars
  • Soil Texture
  • Crop condition surveys
  • Soil surveys (gSSURGO)

We have made our weekly forecast and commentary during the season available publicly on this website. Gro users can access daily forecasts as well as monitor specific inputs to the model (e.g., weekly NDVI updates, daily temperature shifts). For more technical information, you can download our yield model research paper here.