Introduction

Gro’s Yield Forecast Models use a suite of machine-learning models to estimate in-season yields at the district, province, and/or national levels on a daily basis. Gro’s US Corn Yield Forecast Model gives county-level and national-level forecasts. The model uses spatially explicit weather, vegetation health, and soil data, to monitor environmental and crop conditions during the growing season and continuously forecast final yield. 

Customers Use the Model to:

  • Gauge crop availability and crop prices
  • Understand how weather impacts yields in microclimates
  • Inform other models focused on damage caused by pests and diseases
  • Monitor crop production, the most significant part of the balance sheet, by tracking its most variable component, which is in-season yield
  • Make early-season forecasts of end-of-season yield in the US and across the Corn Belt, which produces 80%-85% of national corn

Why It Matters

Corn is a particularly valuable crop to model, given its importance to global food supply and economic trade. The US is the largest producer and consumer of corn, accounting for a third of the world’s corn production. The area planted affects estimates of upcoming harvest sizes, which in turn impact regional and global supply and demand balance sheets. But corn yields can vary substantially across years, with forecast production changing frequently within a season due to rapid shifts in weather.

Methodology

Using USDA data, we applied our agronomic expertise to select the best variables on a county-by-county basis. Then we tested those variables on the 918 most relevant counties, instead of just on one national number. After removing the linear trend growth, we performed a “leave one year out” analysis, for example fitting a model to “forecast” 2011 using every year between 2000 and 2016 except 2011. Some variables selected by our experts include:

  • Normalized difference vegetation index (NDVI)
  • Land surface temperature (LST)
  • Rainfall data (GPM)
  • Crop condition surveys
  • Crop calendars
  • Acreage planted and harvested
  • Soil surveys (gSSURGO)
  • Cropland data (CDL)

Related Analytics