Gro launches new Real-Time Assessment on China’s Food Security

Learn More

Simplified Brazil Soybeans Crop Model: Basic Yield Forecasting Using Gro

Summary & Output

The Brazil soybean crop model offers a simple framework for forecasting yield that uses a relatively small number of inputs and produces reasonably accurate estimates (1% in-sample error rate).
Gro: Crop Weighted NDVI in Brazil
Using historical yield, production quantity, and NDVI data from within Gro, this model develops a weighting of national soybean production by each state, and weighted national NDVI by each state, to build a basic picture of soybean crop health.

Attributes

  • Input your own assumptions and improve the model with your own proprietary data.
  • Use environmental conditions in the model to identify acreage in other countries that will impact production.
Accessibility: The code for Gro’s framework is functional exclusively with the Gro API Client

Inputs

This framework uses several environmental variables including precipitation, temperature, potential evapotranspiration and normalized difference vegetation index (NDVI). Each input source is profiled below.
GIMMS MODIS NDVI

Model Specific Data: 8-day averages for districts of Brazil’s soybean-producing areas

FAO, PS&D and IGC

Model Specific Data: Annual soybean yields at the national level

IBGE and CONAB

Model Specific Data: May be used to determine weighting of each state’s contribution to national soybean production

Methodology

Gro: Brazil Soybean Model Methodology
  • Generate a full history of national-level soybean yield for Brazil
  • Compute the “weight” of each state as the average fraction of soybean production
  • Train a regression model to fit the crop-weighted average of NDVI to national crop yield
  • Based on the results of the previous steps, use current-season NDVI values to forecast yield

Notebook

Contact sales